When your own oversight body, the independent board you created and funded specifically to hold you accountable, publicly says your rules for banning users are inconsistent, opaque, and raise systemic human rights concerns, the problem is no longer a matter of public perception. It is a matter of institutional failure. That is where Meta now stands.&nbsp;The Case That Exposed The System&nbsp;The Oversight Board's review was triggered by a case involving an Instagram account with 70,000 followers that was banned after making threatening posts targeting a journalist. The board upheld the ban itself, the account had violated Meta's rules. But the case opened a wider investigation into how Meta handles account-level enforcement across its platforms, and what the board found was damning.&nbsp;Two Platforms, Two Completely Different Rules&nbsp;Facebook and Instagram, both owned by Meta, both governed by the same parent company, use markedly different penalty structures for the same kinds of violations. Facebook implements a system of temporary suspensions before escalating to a permanent ban, giving users a graduated series of consequences. Instagram does not. Instead, Instagram restricts features like livestreaming access or reduces account visibility as intermediate penalties, but the board pointed out that livestreaming is not even available to all accounts, since it requires a minimum of 1,000 followers. A penalty that restricts a feature most users do not have is not a penalty. It is a gap in the system dressed up as a policy.&nbsp;AI Makes The Decision, Nobody Reviews It&nbsp;The board found that many account bans and suspensions appeared to have been made automatically by AI systems, with no human oversight involved in the decision. Users reported being unable to challenge account disablements or receiving any meaningful explanation for why their accounts were banned. The AI flags an account, the AI disables it, and the user is left with no information about what happened, no human to speak to, and no realistic path to appeal. The board explicitly recommended that users should at minimum be notified when an AI system, rather than a human moderator, has been responsible for penalising their account.&nbsp;750 Public Comments On A Single Case&nbsp;The scale of frustration among Meta's users was reflected in the board's own data. It received more than 750 public comments on this single case alone with endless list of complaint posts (maybe in millions) on other social media platforms like X. Board members also reported receiving what they described as ‘innumerable’ individual complaints from users who had their accounts disabled. For a platform with billions of users, the Oversight Board's comment volume is a small sample, but it signals a level of organised public anger that is difficult to ignore.&nbsp;What The Board Recommended&nbsp;The Oversight Board made several specific recommendations. It called on Meta to provide users with a meaningful appeals process that allows them to submit written explanations, not just click a button and hope. It recommended that Meta notify users when AI systems are responsible for enforcement actions on their accounts. It proposed that information about account bans be included in Meta's public transparency reports. It recommended the creation of a dedicated dashboard where users can review their account status, see their violation history, understand the role AI played in any moderation decisions, and access information about how to appeal. And it called for a dedicated reporting channel for people who are high-risk targets of violence and their representatives.&nbsp;The Bigger Problem&nbsp;Meta created the Oversight Board in 2020 as an independent check on its content moderation decisions. The board was supposed to provide accountability and build public trust. Instead, what the board's latest findings reveal is a company that has automated enforcement at scale, stripped out human review, created inconsistent rules across its own platforms, and left millions of users with no meaningful recourse when the system gets it wrong. The Oversight Board can recommend changes. It cannot force Meta to implement them. Whether any of these recommendations result in real changes to how Meta bans accounts or whether they join the growing list of oversight recommendations that Meta has acknowledged and quietly ignored, remains to be seen.

Meta's Oversight Board found account bans lack due process, AI makes decisions with no human review, and Facebook and Instagram use different rules. 750+ complaints on one case.

When your own oversight body, the independent board you created and funded specifically to hold you accountable, publicly says your rules for banning users are inconsistent, opaque, and raise systemic human rights concerns, the problem is no longer a matter of public perception. It is a matter of institutional failure. That is where Meta now stands. The Case That Exposed The System The Oversight Board's review was triggered by a case involving an Instagram account with 70,000 followers that was banned after making threatening posts targeting a journalist. The board upheld the ban itself, the account had violated Meta's rules. But the case opened a wider investigation into how Meta handles account-level enforcement across its platforms, and what the board found was damning. Two Platforms, Two Completely Different Rules Facebook and Instagram, both owned by Meta, both governed by the same parent company, use markedly different penalty structures for the same kinds of violations. Facebook implements a system of temporary suspensions before escalating to a permanent ban, giving users a graduated series of consequences. Instagram does not. Instead, Instagram restricts features like livestreaming access or reduces account visibility as intermediate penalties, but the board pointed out that livestreaming is not even available to all accounts, since it requires a minimum of 1,000 followers. A penalty that restricts a feature most users do not have is not a penalty. It is a gap in the system dressed up as a policy. AI Makes The Decision, Nobody Reviews It The board found that many account bans and suspensions appeared to have been made automatically by AI systems, with no human oversight involved in the decision. Users reported being unable to challenge account disablements or receiving any meaningful explanation for why their accounts were banned. The AI flags an account, the AI disables it, and the user is left with no information about what happened, no human to speak to, and no realistic path to appeal. The board explicitly recommended that users should at minimum be notified when an AI system, rather than a human moderator, has been responsible for penalising their account. 750 Public Comments On A Single Case The scale of frustration among Meta's users was reflected in the board's own data. It received more than 750 public comments on this single case alone with endless list of complaint posts (maybe in millions) on other social media platforms like X. Board members also reported receiving what they described as ‘innumerable’ individual complaints from users who had their accounts disabled. For a platform with billions of users, the Oversight Board's comment volume is a small sample, but it signals a level of organised public anger that is difficult to ignore. What The Board Recommended The Oversight Board made several specific recommendations. It called on Meta to provide users with a meaningful appeals process that allows them to submit written explanations, not just click a button and hope. It recommended that Meta notify users when AI systems are responsible for enforcement actions on their accounts. It proposed that information about account bans be included in Meta's public transparency reports. It recommended the creation of a dedicated dashboard where users can review their account status, see their violation history, understand the role AI played in any moderation decisions, and access information about how to appeal. And it called for a dedicated reporting channel for people who are high-risk targets of violence and their representatives. The Bigger Problem Meta created the Oversight Board in 2020 as an independent check on its content moderation decisions. The board was supposed to provide accountability and build public trust. Instead, what the board's latest findings reveal is a company that has automated enforcement at scale, stripped out human review, created inconsistent rules across its own platforms, and left millions of users with no meaningful recourse when the system gets it wrong. The Oversight Board can recommend changes. It cannot force Meta to implement them. Whether any of these recommendations result in real changes to how Meta bans accounts or whether they join the growing list of oversight recommendations that Meta has acknowledged and quietly ignored, remains to be seen.

Meta's own independent Oversight Board has found that the company's account banning practices lack due process, transparency, and consistency. Facebook and Instagram use different penalty systems. AI makes decisions with no human oversight. Users cannot meaningfully appeal. The board received over 750 public complaints on a single case — and says its members get ‘innumerable’ individual complaints from banned users.

'Millions of lives destroyed': Meta's oversight board thinks its AI rules for banning accounts are baffling

Meta AI has 1.2 billion monthly active users. That number sounds impressive until you understand how it was achieved: not by building a product people chose to use, but by embedding an AI assistant directly into WhatsApp, Instagram DMs, and Facebook feeds — apps that billions of people were already using every day. Users did not download Meta AI. They did not sign up for it. It appeared, uninvited, inside their existing conversations and search bars. Compare that to ChatGPT, which has 900 million weekly active users who actively chose to visit a standalone product — and which holds 76.85% of the AI search market. For professional tasks like coding, research, legal analysis, and writing, ChatGPT remains far superior. Meta AI is widely described as useful for quick queries and social integration, but not much else.&nbsp;$83.5 Billion In Cumulative Losses And Counting&nbsp;Meta's Reality Labs division — which houses its AI, VR, AR, and metaverse operations — has accumulated approximately $83.5 billion in operating losses across 21 consecutive quarters. In 2025 alone, Reality Labs burned $19.2 billion. In Q1 2026, it lost another $4.03 billion while generating just $402 million in revenue — a ratio of roughly $10 lost for every $1 earned. Meta has now laid off more than 1,000 Reality Labs employees as it pivots resources from the metaverse to AI, but the losses continue to mount with no breakeven point in sight.&nbsp;$145 Billion In AI Spending — With No Cloud Business To Show For It&nbsp;Meta has raised its 2026 capital expenditure forecast to between $125 billion and $145 billion — an increase of over 60% year-on-year. This money is going into data centres, Nvidia GPUs, and AI infrastructure. Google, Amazon, and Microsoft are spending similar amounts — but each of them has a cloud computing business that immediately monetises that infrastructure by selling compute capacity to other companies. Meta has no cloud business. Its revenue is almost entirely advertising. Every dollar spent on AI infrastructure is a bet that it will eventually make ads more targeted and platforms more engaging. That may happen. But right now, Meta AI, business agents, WhatsApp paid messaging, and early commerce initiatives are not generating meaningful revenue. The AI spending is improving the existing ad business — not building a new one.&nbsp;The Moderation Crisis Nobody Is Pricing In&nbsp;While Meta pours billions into AI infrastructure, the AI it has already deployed is actively destroying trust with the people who use its platforms. Meta's own Oversight Board confirmed in June 2026 that the company's account bans lack due process and transparency. Thousands of creators, small business owners, and ordinary users have had their accounts suspended or permanently deleted by AI moderation systems that falsely flag normal content — including family photos — as violations. A Change.org petition against wrongful Meta account bans has gathered nearly 56,000 signatures. Meanwhile, AI-generated bot accounts with blue tick verification badges continue to operate freely on the same platforms. Meta has gutted its human moderation teams in favour of AI enforcement — and the AI is getting it catastrophically wrong at scale.&nbsp;Zuckerberg's Answer On AI Returns&nbsp;When Meta reported its Q1 2026 earnings and revealed the raised capex guidance, analysts pressed Mark Zuckerberg on what measurable return on investment the company's AI spending was actually delivering. His answer was notable: he described it as ‘a very technical question.’ For a company spending up to $145 billion in a single year on capital projects — more than the GDP of most countries — that answer is not reassuring. Meta's CFO said the 2025 AI investment cohort ‘paid off,’ but what that means in practice is that ad targeting improved. It does not mean Meta AI as a product generates revenue. It does not mean the $145 billion in infrastructure will produce a new business line. It means the existing advertising machine got slightly better.&nbsp;What The Market Is Choosing To Ignore&nbsp;Meta's stock has fallen approximately 27.5% from its all-time high of $796.25 in August 2025, trading at around $577 as of mid-June 2026. But even after that correction, the market is still valuing Meta at roughly $1.46 trillion — a price that assumes the AI spending will eventually pay off, that Meta AI will become a competitive product, and that the moderation crisis will resolve itself. The data suggests a different picture: a company that has burned nearly $90 billion on Reality Labs over seven years with no return, that is spending $145 billion on infrastructure it cannot monetise through a cloud business, whose flagship AI product was force-embedded into existing apps rather than adopted on its own merits, and whose AI moderation system is actively destroying the creator and small business ecosystem that drives its advertising revenue. The market may be right that Meta's ad business is strong enough to survive all of this. But calling Meta AI a success — when its users did not choose it, its moderation is failing, its losses are approaching $90 billion, and its CEO calls ROI ‘a very technical question’ — requires ignoring a substantial amount of evidence to the contrary.

Meta AI has 1.2B users it didn't earn, $83.5B in cumulative losses, $145B capex with no cloud business, and a moderation crisis destroying creator trust. Here's the data.

Meta AI has 1.2 billion monthly active users. That number sounds impressive until you understand how it was achieved: not by building a product people chose to use, but by embedding an AI assistant directly into WhatsApp, Instagram DMs, and Facebook feeds — apps that billions of people were already using every day. Users did not download Meta AI. They did not sign up for it. It appeared, uninvited, inside their existing conversations and search bars. Compare that to ChatGPT, which has 900 million weekly active users who actively chose to visit a standalone product — and which holds 76.85% of the AI search market. For professional tasks like coding, research, legal analysis, and writing, ChatGPT remains far superior. Meta AI is widely described as useful for quick queries and social integration, but not much else. $83.5 Billion In Cumulative Losses And Counting Meta's Reality Labs division — which houses its AI, VR, AR, and metaverse operations — has accumulated approximately $83.5 billion in operating losses across 21 consecutive quarters. In 2025 alone, Reality Labs burned $19.2 billion. In Q1 2026, it lost another $4.03 billion while generating just $402 million in revenue — a ratio of roughly $10 lost for every $1 earned. Meta has now laid off more than 1,000 Reality Labs employees as it pivots resources from the metaverse to AI, but the losses continue to mount with no breakeven point in sight. $145 Billion In AI Spending — With No Cloud Business To Show For It Meta has raised its 2026 capital expenditure forecast to between $125 billion and $145 billion — an increase of over 60% year-on-year. This money is going into data centres, Nvidia GPUs, and AI infrastructure. Google, Amazon, and Microsoft are spending similar amounts — but each of them has a cloud computing business that immediately monetises that infrastructure by selling compute capacity to other companies. Meta has no cloud business. Its revenue is almost entirely advertising. Every dollar spent on AI infrastructure is a bet that it will eventually make ads more targeted and platforms more engaging. That may happen. But right now, Meta AI, business agents, WhatsApp paid messaging, and early commerce initiatives are not generating meaningful revenue. The AI spending is improving the existing ad business — not building a new one. The Moderation Crisis Nobody Is Pricing In While Meta pours billions into AI infrastructure, the AI it has already deployed is actively destroying trust with the people who use its platforms. Meta's own Oversight Board confirmed in June 2026 that the company's account bans lack due process and transparency. Thousands of creators, small business owners, and ordinary users have had their accounts suspended or permanently deleted by AI moderation systems that falsely flag normal content — including family photos — as violations. A Change.org petition against wrongful Meta account bans has gathered nearly 56,000 signatures. Meanwhile, AI-generated bot accounts with blue tick verification badges continue to operate freely on the same platforms. Meta has gutted its human moderation teams in favour of AI enforcement — and the AI is getting it catastrophically wrong at scale. Zuckerberg's Answer On AI Returns When Meta reported its Q1 2026 earnings and revealed the raised capex guidance, analysts pressed Mark Zuckerberg on what measurable return on investment the company's AI spending was actually delivering. His answer was notable: he described it as ‘a very technical question.’ For a company spending up to $145 billion in a single year on capital projects — more than the GDP of most countries — that answer is not reassuring. Meta's CFO said the 2025 AI investment cohort ‘paid off,’ but what that means in practice is that ad targeting improved. It does not mean Meta AI as a product generates revenue. It does not mean the $145 billion in infrastructure will produce a new business line. It means the existing advertising machine got slightly better. What The Market Is Choosing To Ignore Meta's stock has fallen approximately 27.5% from its all-time high of $796.25 in August 2025, trading at around $577 as of mid-June 2026. But even after that correction, the market is still valuing Meta at roughly $1.46 trillion — a price that assumes the AI spending will eventually pay off, that Meta AI will become a competitive product, and that the moderation crisis will resolve itself. The data suggests a different picture: a company that has burned nearly $90 billion on Reality Labs over seven years with no return, that is spending $145 billion on infrastructure it cannot monetise through a cloud business, whose flagship AI product was force-embedded into existing apps rather than adopted on its own merits, and whose AI moderation system is actively destroying the creator and small business ecosystem that drives its advertising revenue. The market may be right that Meta's ad business is strong enough to survive all of this. But calling Meta AI a success — when its users did not choose it, its moderation is failing, its losses are approaching $90 billion, and its CEO calls ROI ‘a very technical question’ — requires ignoring a substantial amount of evidence to the contrary.

Meta AI has 1.2 billion monthly users it did not earn, force-embedded into WhatsApp and Instagram. Reality Labs has burned through $83.5 billion in cumulative losses. The company is spending $145 billion on AI infrastructure with no cloud business to monetise it. Its own Oversight Board says its AI moderation lacks due process. And when asked about AI returns, Mark Zuckerberg called it ‘a very technical question.’ Here is the data the market is choosing to ignore.

Meta AI is a lost cause but the market refuses to see it yet! Here's why

Meta Platforms has agreed to invest $900 million in Indian fintech startup CRED, acquiring an approximately 20 per cent minority stake and valuing the Bengaluru-based company at $4.5 billion. At the same time, CRED founder Kunal Shah has been appointed global head of WhatsApp, succeeding Will Cathcart, who led the platform for nearly seven years.The deal is the single largest strategic investment by a global technology major in India's consumer financial services sector. But the number that matters most is not $900 million. It is 0.65 per cent.<h2>What Meta actually bought</h2>CRED is not a platform built for everyone, and it was never meant to be.The company admits only Indians with CIBIL credit scores above 750, a threshold that excludes the vast majority of the population and leaves roughly 17 million members who collectively represent the country's most financially active consumer segment. These are people who pay their credit card bills on time, borrow responsibly, and spend at a scale that makes them disproportionately valuable to any financial platform trying to build depth rather than just volume.The platform processes more than 40 per cent of India's credit card bill payments. Over time, it has expanded into personal loans, insurance, e-commerce and investment products. In March 2026, the Reserve Bank of India granted CRED a Payment Aggregator licence, one of the most consequential regulatory authorisations in Indian fintech, giving it the infrastructure to process payments directly for merchants.Financially, CRED ended FY25 with operating revenue of Rs 2,735 crore, up 16 per cent year on year. Operating losses narrowed by 51 per cent to Rs 298 crore. On a net basis, after accounting for non-operating expenses, including employee stock options and depreciation, losses contracted by 11.5 per cent to Rs 1,457 crore. In May 2026 alone, CRED processed 157.19 crore UPI transactions worth Rs 61,002 crore.Ashish Bhatia, Founder and CEO of India Accelerator and Co-Founder of Finvolve, said the deal should not be read as a valuation story."People often say CRED is a credit card bill payment app. That's like saying Amazon is a bookstore," Bhatia said. "CRED became valuable because it became the platform for a particular kind of Indian: financially responsible, creditworthy and aspirational, allowing them to progressively do more things. Pay bills, yes. But also borrow, invest, shop and discover. The product kept deepening because the relationship kept deepening. That's what platforms do. That's what Meta bought."<h2>The gap Meta could not close</h2>The strategic urgency behind the deal is visible in that single number: 0.65 per cent.That is WhatsApp Pay's share of India's UPI market as of May 2026, according to data published by the National Payments Corporation of India and reported across multiple outlets. It trails PhonePe at 46.2 per cent, Google Pay at 32.7 per cent, Paytm at 7.9 per cent, and even smaller players such as Navi, super.money, BHIM, FamPay and CRED's own UPI product. For an app used by more than 500 million Indians every day, it is a number that tells its own story.India processed a record 23.2 billion UPI transactions in May 2026 alone, according to NPCI. WhatsApp was present for almost none of them in any meaningful commercial sense.The reason traces back to timing. PhonePe entered the UPI market in 2016. Google Pay followed in 2017. WhatsApp Pay did not receive full regulatory approval until 2020, by which time Indian consumers had already formed payment habits, merchants had already integrated rival platforms, and the market had already chosen its defaults. Meta has spent six years trying to close that gap. It has not.Meta has now committed more than $6.6 billion across India's digital economy: $5.7 billion for a 9.99 per cent stake in Reliance's Jio Platforms in April 2020, and $900 million for an approximately 20 per cent stake in CRED in June 2026. The Jio investment was designed to accelerate WhatsApp commerce through JioMart. That partnership remains active. WhatsApp Pay's market position, however, has not materially improved.<h2>One window, one deadline</h2>The window Meta is now targeting is not competitive. It is regulatory, and it has a closing date.India's National Payments Corporation of India requires PhonePe and Google Pay to cap their individual transaction volumes at 30 per cent of the total UPI market by 31 December 2026. Both currently exceed that threshold significantly, with PhonePe at 46.2 per cent and Google Pay at 32.7 per cent. The rule has been extended twice before, most recently on 31 December 2024, citing operational challenges. If it remains in force this time, both dominant platforms would be required to stop onboarding new users once they reach the cap, creating an opportunity that WhatsApp Pay, with Kunal Shah now leading the product globally, is better positioned than ever to seize.The shift is already beginning. The combined share of PhonePe and Google Pay fell to 79 per cent in May 2026, below 80 per cent for the first time since NPCI began publishing individual application statistics, according to Outlook Business, citing NPCI data. The top three providers collectively held 95.2 per cent of all transactions in January 2024. By May 2026, that had contracted to 87 per cent as smaller platforms steadily gained ground.NPCI also lifted the user onboarding cap on WhatsApp Pay entirely on 31 December 2024, opening UPI payments on the platform to its full Indian user base for the first time.<h2>What this means for the ecosystem</h2>The deal's implications extend well beyond WhatsApp and CRED.Bhatia said it is a signal that India's fintech market is entering a period of fundamental bifurcation and that the criteria for survival are shifting."The ecosystem is bifurcating sharply between companies that own a real relationship with a specific community and companies that are just running transactions," he said. "The ones who survive the next cycle will be asking the right question today: not 'How do I grow faster?' but 'Which community do I truly belong to, and what more can I do for them?'"The CRED valuation trajectory illustrates this shift in numbers. The company peaked at $6.4 billion in 2022, fell to $3.5 billion in a 2025 down round, and has recovered to $4.5 billion following Meta's investment. It is a full cycle that mirrors the broader shift in sentiment across India's fintech sector over the past three years. Community depth, rather than transaction volume, is what preserved its value through that correction.<h2>The next layer: intelligence, not more rails</h2>For founders watching this deal, Paramdeep Singh, Founder of Long Tail Ventures, said the real signal has nothing to do with building another payments product."India already processes over 18 billion UPI transactions every month. The next opportunity is no longer building another payments app, but building intelligence on top of existing financial rails," Singh said.Singh pointed to underwriting, fraud detection, collections, compliance, treasury management and AI-powered financial tools for consumers and businesses as the most significant opportunities emerging from India's digital financial infrastructure."These are large, recurring problems where AI can materially improve cost, speed and accuracy," he said. "India's digital public infrastructure provides a unique foundation to build globally relevant companies in these areas. For founders, the challenge is no longer access to financial rails, but finding proprietary data and distribution that others cannot easily replicate."The scale of that infrastructure is not in dispute. NPCI data show India's UPI network processed 228.3 billion transactions during 2025, up 29.3 per cent by volume year on year. The rails are built, and the question now is who builds the intelligence layer on top of them.<h2>The data question regulators will watch</h2>The deal does raise a question that India's regulatory architecture will need to answer over time.CRED holds one of the richest repositories of consumer financial behaviour data in India, including credit card repayment histories, spending patterns, borrowing behaviour and UPI transaction data, all tied to the country's most creditworthy consumer segment. Meta, as a minority investor without a board seat, has said it will have no access to that member data, a structural choice designed to address the data sovereignty concerns that have historically made Indian regulators cautious about foreign technology companies expanding into financial services.India's Digital Personal Data Protection Act 2023 is the governing framework for how platforms handle consumer data of this nature. Its rules covering consent managers and Significant Data Fiduciaries are still being operationalised. Whether the current architecture of the Meta-CRED deal satisfies the RBI and NPCI as WhatsApp's financial services ambitions continue to grow will be a question that regulators, and the market, will be watching closely.

With WhatsApp Pay holding less than 1% of India's UPI market despite 500 million users, the deal of  $900 million investment by Meta in CRED is less about money and more about a payments problem Meta has spent six years failing to solve.

Meta Platforms has agreed to invest $900 million in Indian fintech startup CRED, acquiring an approximately 20 per cent minority stake and valuing the Bengaluru-based company at $4.5 billion. At the same time, CRED founder Kunal Shah has been appointed global head of WhatsApp, succeeding Will Cathcart, who led the platform for nearly seven years. The deal is the single largest strategic investment by a global technology major in India's consumer financial services sector. But the number that matters most is not $900 million. It is 0.65 per cent. What Meta actually bought CRED is not a platform built for everyone, and it was never meant to be. The company admits only Indians with CIBIL credit scores above 750, a threshold that excludes the vast majority of the population and leaves roughly 17 million members who collectively represent the country's most financially active consumer segment. These are people who pay their credit card bills on time, borrow responsibly, and spend at a scale that makes them disproportionately valuable to any financial platform trying to build depth rather than just volume. The platform processes more than 40 per cent of India's credit card bill payments. Over time, it has expanded into personal loans, insurance, e-commerce and investment products. In March 2026, the Reserve Bank of India granted CRED a Payment Aggregator licence, one of the most consequential regulatory authorisations in Indian fintech, giving it the infrastructure to process payments directly for merchants. Financially, CRED ended FY25 with operating revenue of Rs 2,735 crore, up 16 per cent year on year. Operating losses narrowed by 51 per cent to Rs 298 crore. On a net basis, after accounting for non-operating expenses, including employee stock options and depreciation, losses contracted by 11.5 per cent to Rs 1,457 crore. In May 2026 alone, CRED processed 157.19 crore UPI transactions worth Rs 61,002 crore. Ashish Bhatia, Founder and CEO of India Accelerator and Co-Founder of Finvolve, said the deal should not be read as a valuation story. "People often say CRED is a credit card bill payment app. That's like saying Amazon is a bookstore," Bhatia said. "CRED became valuable because it became the platform for a particular kind of Indian: financially responsible, creditworthy and aspirational, allowing them to progressively do more things. Pay bills, yes. But also borrow, invest, shop and discover. The product kept deepening because the relationship kept deepening. That's what platforms do. That's what Meta bought." The gap Meta could not close The strategic urgency behind the deal is visible in that single number: 0.65 per cent. That is WhatsApp Pay's share of India's UPI market as of May 2026, according to data published by the National Payments Corporation of India and reported across multiple outlets. It trails PhonePe at 46.2 per cent, Google Pay at 32.7 per cent, Paytm at 7.9 per cent, and even smaller players such as Navi, super.money, BHIM, FamPay and CRED's own UPI product. For an app used by more than 500 million Indians every day, it is a number that tells its own story. India processed a record 23.2 billion UPI transactions in May 2026 alone, according to NPCI. WhatsApp was present for almost none of them in any meaningful commercial sense. The reason traces back to timing. PhonePe entered the UPI market in 2016. Google Pay followed in 2017. WhatsApp Pay did not receive full regulatory approval until 2020, by which time Indian consumers had already formed payment habits, merchants had already integrated rival platforms, and the market had already chosen its defaults. Meta has spent six years trying to close that gap. It has not. Meta has now committed more than $6.6 billion across India's digital economy: $5.7 billion for a 9.99 per cent stake in Reliance's Jio Platforms in April 2020, and $900 million for an approximately 20 per cent stake in CRED in June 2026. The Jio investment was designed to accelerate WhatsApp commerce through JioMart. That partnership remains active. WhatsApp Pay's market position, however, has not materially improved. One window, one deadline The window Meta is now targeting is not competitive. It is regulatory, and it has a closing date. India's National Payments Corporation of India requires PhonePe and Google Pay to cap their individual transaction volumes at 30 per cent of the total UPI market by 31 December 2026. Both currently exceed that threshold significantly, with PhonePe at 46.2 per cent and Google Pay at 32.7 per cent. The rule has been extended twice before, most recently on 31 December 2024, citing operational challenges. If it remains in force this time, both dominant platforms would be required to stop onboarding new users once they reach the cap, creating an opportunity that WhatsApp Pay, with Kunal Shah now leading the product globally, is better positioned than ever to seize. The shift is already beginning. The combined share of PhonePe and Google Pay fell to 79 per cent in May 2026, below 80 per cent for the first time since NPCI began publishing individual application statistics, according to Outlook Business, citing NPCI data. The top three providers collectively held 95.2 per cent of all transactions in January 2024. By May 2026, that had contracted to 87 per cent as smaller platforms steadily gained ground. NPCI also lifted the user onboarding cap on WhatsApp Pay entirely on 31 December 2024, opening UPI payments on the platform to its full Indian user base for the first time. What this means for the ecosystem The deal's implications extend well beyond WhatsApp and CRED. Bhatia said it is a signal that India's fintech market is entering a period of fundamental bifurcation and that the criteria for survival are shifting. "The ecosystem is bifurcating sharply between companies that own a real relationship with a specific community and companies that are just running transactions," he said. "The ones who survive the next cycle will be asking the right question today: not 'How do I grow faster?' but 'Which community do I truly belong to, and what more can I do for them?'" The CRED valuation trajectory illustrates this shift in numbers. The company peaked at $6.4 billion in 2022, fell to $3.5 billion in a 2025 down round, and has recovered to $4.5 billion following Meta's investment. It is a full cycle that mirrors the broader shift in sentiment across India's fintech sector over the past three years. Community depth, rather than transaction volume, is what preserved its value through that correction. The next layer: intelligence, not more rails For founders watching this deal, Paramdeep Singh, Founder of Long Tail Ventures, said the real signal has nothing to do with building another payments product. "India already processes over 18 billion UPI transactions every month. The next opportunity is no longer building another payments app, but building intelligence on top of existing financial rails," Singh said. Singh pointed to underwriting, fraud detection, collections, compliance, treasury management and AI-powered financial tools for consumers and businesses as the most significant opportunities emerging from India's digital financial infrastructure. "These are large, recurring problems where AI can materially improve cost, speed and accuracy," he said. "India's digital public infrastructure provides a unique foundation to build globally relevant companies in these areas. For founders, the challenge is no longer access to financial rails, but finding proprietary data and distribution that others cannot easily replicate." The scale of that infrastructure is not in dispute. NPCI data show India's UPI network processed 228.3 billion transactions during 2025, up 29.3 per cent by volume year on year. The rails are built, and the question now is who builds the intelligence layer on top of them. The data question regulators will watch The deal does raise a question that India's regulatory architecture will need to answer over time. CRED holds one of the richest repositories of consumer financial behaviour data in India, including credit card repayment histories, spending patterns, borrowing behaviour and UPI transaction data, all tied to the country's most creditworthy consumer segment. Meta, as a minority investor without a board seat, has said it will have no access to that member data, a structural choice designed to address the data sovereignty concerns that have historically made Indian regulators cautious about foreign technology companies expanding into financial services. India's Digital Personal Data Protection Act 2023 is the governing framework for how platforms handle consumer data of this nature. Its rules covering consent managers and Significant Data Fiduciaries are still being operationalised. Whether the current architecture of the Meta-CRED deal satisfies the RBI and NPCI as WhatsApp's financial services ambitions continue to grow will be a question that regulators, and the market, will be watching closely.

With WhatsApp Pay holding less than 1% of India's UPI market despite 500 million users, the deal of &nbsp;$900 million investment by Meta in CRED is less about money and more about a payments problem Meta has spent six years failing to solve.

With $900 million, Meta eyes at 17 million members, one regulatory window and a payments problem in India

Meta is making a new push into wearable artificial intelligence. On Tuesday, they unveiled a new line of AI-powered smart glasses developed with eyewear giant EssilorLuxottica. Starting at $299, the new devices are considerably cheaper than Meta's premium smart glasses launched last year.&nbsp;The announcement is important because it signals Meta's ambition to make AI hardware more accessible to everyday consumers as competition in the wearable technology market intensifies. The launch also comes as rivals including Apple, Google and Snap continue investing heavily in next-generation smart glasses and AI-powered devices.<h2>Meta is betting big on AI glasses</h2>For several years, Meta has argued that smart glasses could become one of the most important ways people interact with artificial intelligence. Meta’s latest glasses are powered by Meta AI running on Muse Spark, the first model developed by Meta's Superintelligence Labs. According to Meta, the glasses allow users to interact with AI, access information and receive text-based assistance through a wearable device rather than a smartphone. The launch reflects CEO Mark Zuckerberg's broader vision of bringing what he calls "personal intelligence" directly to consumers.<figure class="image"><img src="https://about.fb.com/wp-content/uploads/2026/06/02_A-Pair-for-Any-Look_Carousel-02.jpg?fit=1920%2C1672" alt="Meta glasses"></figure><h2>What's different about the new glasses?</h2>One notable change is branding. Unlike previous devices, the new glasses are not linked to popular eyewear brands such as Ray-Ban or Oakley. Instead, Meta is introducing its own design language with several new frame styles and colours. Among the options is a slim oval collection developed in collaboration with media personality Kylie Jenner. The company hopes the broader range of styles will help attract consumers who may not have been interested in earlier smart glasses. A growing market<figure class="image"><img src="https://about.fb.com/wp-content/uploads/2026/06/02_A-Pair-for-Any-Look_Carousel-03.jpg?fit=1920%2C1672" alt="Meta glasses"></figure><h2><img src="https://about.fb.com/wp-content/uploads/2026/06/02_A-Pair-for-Any-Look_Carousel-01.jpg?fit=1920%2C1672" alt="Meta glasses">Meta enters the launch from a position of strength</h2>According to data from the International Data Corporation (IDC), global smart-glasses shipments reached 9.6 million units last year. Meta accounted for approximately 76.1 per cent of those shipments. That dominance has encouraged competitors to accelerate their own wearable AI projects. Google is expanding its smart-glasses efforts, while Apple is reportedly developing future wearable products that could incorporate artificial intelligence. Meanwhile, Snap recently launched augmented reality glasses priced at $2,195, targeting a different segment of the market.<h2>Why the lower price matters</h2>The biggest talking point may be the price. At $299, the new Meta Glasses cost significantly less than last year's $800 Ray-Ban Display glasses. Industry analysts have long argued that price remains one of the biggest barriers preventing smart glasses from becoming mainstream. By lowering the entry cost, Meta appears to be prioritising adoption and ecosystem growth over premium positioning.&nbsp;The next phase of the AI race.Artificial intelligence is increasingly moving beyond chatbots and smartphones. Technology companies are now competing to determine which device will become the primary way people interact with AI throughout the day.<h2>Meta believes smart glasses could play that role</h2>Whether consumers agree remains to be seen. However, with a lower price, new designs and the company's latest AI technology, Meta is making one thing clear: it sees wearable AI as one of the most important battlegrounds in the technology industry.

Meta has launched a new range of AI-powered smart glasses starting at $299, significantly undercutting previous models. Powered by Meta's latest AI model.

Meta is making a new push into wearable artificial intelligence. On Tuesday, they unveiled a new line of AI-powered smart glasses developed with eyewear giant EssilorLuxottica. Starting at $299, the new devices are considerably cheaper than Meta's premium smart glasses launched last year. The announcement is important because it signals Meta's ambition to make AI hardware more accessible to everyday consumers as competition in the wearable technology market intensifies. The launch also comes as rivals including Apple, Google and Snap continue investing heavily in next-generation smart glasses and AI-powered devices. Meta is betting big on AI glasses For several years, Meta has argued that smart glasses could become one of the most important ways people interact with artificial intelligence. Meta’s latest glasses are powered by Meta AI running on Muse Spark, the first model developed by Meta's Superintelligence Labs. According to Meta, the glasses allow users to interact with AI, access information and receive text-based assistance through a wearable device rather than a smartphone. The launch reflects CEO Mark Zuckerberg's broader vision of bringing what he calls "personal intelligence" directly to consumers. What's different about the new glasses? One notable change is branding. Unlike previous devices, the new glasses are not linked to popular eyewear brands such as Ray-Ban or Oakley. Instead, Meta is introducing its own design language with several new frame styles and colours. Among the options is a slim oval collection developed in collaboration with media personality Kylie Jenner. The company hopes the broader range of styles will help attract consumers who may not have been interested in earlier smart glasses. A growing market Meta enters the launch from a position of strength According to data from the International Data Corporation (IDC), global smart-glasses shipments reached 9.6 million units last year. Meta accounted for approximately 76.1 per cent of those shipments. That dominance has encouraged competitors to accelerate their own wearable AI projects. Google is expanding its smart-glasses efforts, while Apple is reportedly developing future wearable products that could incorporate artificial intelligence. Meanwhile, Snap recently launched augmented reality glasses priced at $2,195, targeting a different segment of the market. Why the lower price matters The biggest talking point may be the price. At $299, the new Meta Glasses cost significantly less than last year's $800 Ray-Ban Display glasses. Industry analysts have long argued that price remains one of the biggest barriers preventing smart glasses from becoming mainstream. By lowering the entry cost, Meta appears to be prioritising adoption and ecosystem growth over premium positioning. The next phase of the AI race.Artificial intelligence is increasingly moving beyond chatbots and smartphones. Technology companies are now competing to determine which device will become the primary way people interact with AI throughout the day. Meta believes smart glasses could play that role Whether consumers agree remains to be seen. However, with a lower price, new designs and the company's latest AI technology, Meta is making one thing clear: it sees wearable AI as one of the most important battlegrounds in the technology industry.

Meta has launched a new range of AI-powered smart glasses starting at $299, significantly undercutting previous models. Powered by Meta's latest AI model.

Meta introduces new AI smart glasses powered by its latest AI model, starting at $299

Meta is reportedly exploring a major investment in Indian fintech startup CRED, with discussions valuing the company at around $4 billion. Reports suggest the deal could involve tens of millions of dollars and have fueled speculation about a deeper partnership—or even a larger strategic move involving CRED founder Kunal Shah. At the same time, Meta’s leadership decisions around its messaging ecosystem have intensified focus on the company’s growing ambitions in India, one of its most important markets worldwide.


Meta's India Bet: Meta Taps Indian Founder To Lead WhatsApp

Facebook’s parent company Meta is investing around $900 million (around Rs 8,550 crore) in Indian fintech startup CRED, the two companies announced on Monday. As part of the investment, Cred’s founder, Kunal Shah, will take over as global CEO of WhatsApp.&nbsp;CRED will now be led by Miten Sampat as the interim CEO, shared Kunal Shah in a Monday post on X.&nbsp;&nbsp;“Kunal Shah will join Meta as WhatsApp’s next leader. Kunal built CRED into one of India’s most important technology companies, and he brings the kind of builder mentality and global perspective that will serve him well in running the world’s biggest messaging app. I look forward to working with Kunal to continue to make WhatsApp the best service for billions of people and millions of businesses,” Mark Zuckerberg, CEO of Meta, said on Monday.&nbsp;Also Read: <a href="https://www.wionews.com/india-news/14-students-killed-in-fire-at-animation-centre-in-lucknow-many-jump-from-1st-floor-1782130828955">15 killed in fire at animation centre in Lucknow; many jump off 1st floor</a>&nbsp;Shah takes over from WhatApp’s current head, Will Cathcart, who will be stepping down after seven years in the role.&nbsp;&nbsp;“Will’s been one of Meta’s most important and effective leaders, helping to bring WhatsApp to over 3 billion people and championing privacy for our community. Will is transitioning to a new role within Meta where he’ll build new products from the ground-up,” Zuckerberg added.&nbsp;&nbsp;Also Read: <a href="https://www.wionews.com/india-news/6-shiv-sena-ubt-rebel-mps-join-eknath-shinde-s-shiv-sena-1782129323513">6 Shiv Sena (UBT) rebel MPs join Eknath Shinde's Shiv Sena</a>

<blockquote class="twitter-tweet">It’s been a minute. 2015–2018 - Exited FreeCharge. Spent time learning and investing. - Pondered about: Why can&#39;t trust be rewarded? Started with $1M of personal capital. - Launched CRED to reward people for paying credit card bills on time. 2019–2025 - Built a system run by a…&mdash; Kunal Shah (@kunalb11) <a href="https://x.com/kunalb11/status/2069043017420599323?ref_src=twsrc%5Etfw">June 22, 2026</a></blockquote> <script async src="https://platform.x.com/widgets.js" charset="utf-8"></script>

&nbsp;&nbsp;&nbsp;Shah said in a post on X, “I started Cred in 2018 with a belief that creditworthiness deserves to be rewarded. In under eight years, that belief has turned into a new category: millions of members, ~₹3,200 crore (~US $325M) in revenue, profitability, a full stack of licences and a strong brand. On this foundation, with additional capital and an extraordinarily talented team, CRED is poised to become an enduring institution for decades to come. I’m stepping back with gratitude and with conviction that the team will keep raising the bar.”&nbsp;Also Read: <a href="https://www.wionews.com/world/india-to-sell-supersonic-brahmos-missile-to-uae-talks-in-its-initial-stages-report-1782133327073">India to sell supersonic BrahMos missile to UAE, talks in its initial stages: Report</a>&nbsp;“We have a generational opportunity to build on Kunal’s vision and compound consistently towards becoming a public company. I’m excited to take CRED forward in its next chapter. We are just getting started,” said Miten Sampat.

Meta CEO Mark Zuckerberg said on Monday that CRED founder Kunal Shah will join Meta as WhatsApp’s next global CEO. Kunal brings the kind of builder mentality and global perspective that will serve him well in running the world’s biggest messaging app, he added.

Facebook’s parent company Meta is investing around $900 million (around Rs 8,550 crore) in Indian fintech startup CRED, the two companies announced on Monday. As part of the investment, Cred’s founder, Kunal Shah, will take over as global CEO of WhatsApp. CRED will now be led by Miten Sampat as the interim CEO, shared Kunal Shah in a Monday post on X. “Kunal Shah will join Meta as WhatsApp’s next leader. Kunal built CRED into one of India’s most important technology companies, and he brings the kind of builder mentality and global perspective that will serve him well in running the world’s biggest messaging app. I look forward to working with Kunal to continue to make WhatsApp the best service for billions of people and millions of businesses,” Mark Zuckerberg, CEO of Meta, said on Monday. Also Read: 15 killed in fire at animation centre in Lucknow; many jump off 1st floor Shah takes over from WhatApp’s current head, Will Cathcart, who will be stepping down after seven years in the role. “Will’s been one of Meta’s most important and effective leaders, helping to bring WhatsApp to over 3 billion people and championing privacy for our community. Will is transitioning to a new role within Meta where he’ll build new products from the ground-up,” Zuckerberg added. Also Read: 6 Shiv Sena (UBT) rebel MPs join Eknath Shinde's Shiv Sena

Meta CEO Mark Zuckerberg said on Monday that CRED founder Kunal Shah will join Meta as WhatsApp’s next global CEO. Kunal brings the kind of builder mentality and global perspective that will serve him well in running the world’s biggest messaging app, he added.

CRED founder Kunal Shah becomes global CEO of WhatsApp as Meta invests $900 million in startup

Meta has accused Israeli spyware firm NSO Group of violating a court injunction in an ongoing legal battle, escalating tensions between the tech giant and the surveillance company. The dispute centers around allegations of unauthorized access and misuse of WhatsApp systems through sophisticated spyware tools.

Meta Accuses NSO of Violating Court  Injunction  | WhatsApp Disrupts Phishing Attempts

Thailand’s Consumer Council is planning a civil lawsuit against Meta over alleged failures to tackle online scams and harmful content on its platforms. Officials say consumers have suffered financial losses linked to fraudulent advertisements and deceptive activity circulating online. The move adds to growing global scrutiny of major tech companies over user safety and platform accountability.


Thailand's Consumer Council Plans Civil Suit Against Meta

The European Union’s top court has upheld Messenger’s “gatekeeper” designation under the bloc’s landmark digital competition rules. The ruling strengthens the EU’s push to regulate major tech platforms and ensure fair competition in the digital market. The decision could have significant implications for how large technology companies operate, share data, and comply with strict European regulations. Watch the latest updates on the court ruling, Big Tech regulation, and the future of digital markets in Europe.


EU Court Upholds Messenger's 'Gatekeeper' Designation 

Meta is reportedly working on a new artificial intelligence-powered wearable device that could take the form of a pendant worn around the neck. According to a report by The Information, the company plans to begin testing the device within the next year as it continues expanding its AI hardware strategy.The reported project comes as major technology companies increasingly explore AI-powered devices beyond smartphones and computers. Meta has already invested heavily in smart glasses and virtual reality products, and the new pendant could become another part of its growing wearable ecosystem.<h2>What is Meta's AI pendant?</h2>According to reports, Meta's new device would function as an AI-powered pendant that users could wear around their neck or attach to clothing. While full details have not been revealed, the wearable is expected to build on technology developed by Limitless, an AI startup acquired by Meta in late 2025.Before the acquisition, Limitless developed a wearable pendant capable of recording conversations, creating summaries and helping users recall information from meetings and daily interactions. Following the acquisition, Meta said the deal would help accelerate its efforts to develop AI-enabled wearable technology. Industry observers believe Meta could integrate its AI assistant into the device, allowing users to interact with artificial intelligence throughout the day without needing to use a smartphone.<h2>Why is Meta investing in AI wearables?</h2>The move reflects a broader trend across the technology industry.Companies are increasingly exploring devices that allow users to access AI more naturally throughout the day. Instead of opening an app or typing a question, wearable devices can provide hands-free access to AI assistants.Meta CEO Mark Zuckerberg has repeatedly argued that wearable devices could become one of the most important ways people interact with artificial intelligence in the future. The company has already found success with its smart glasses developed in partnership with EssilorLuxottica, which combine cameras, audio features and AI-powered assistance. An AI pendant would represent another step towards making AI available throughout the day in a less intrusive form.<h2>Why have AI wearables struggled in the past?</h2>Despite growing interest in artificial intelligence, many dedicated AI wearable devices have struggled to gain widespread adoption. Several products launched over the past few years faced challenges related to privacy concerns, battery life, limited functionality and unclear consumer demand. Some users questioned whether wearable devices that continuously listen to conversations or collect information could create privacy risks. Others found that early AI devices did not offer enough practical benefits compared with smartphones. These challenges have led some technology companies to rethink how AI hardware should be designed and marketed. Meta appears to be betting that advances in AI models and wearable technology could make a new generation of devices more useful.<h2>Meta's broader wearable strategy</h2>The reported pendant project is only one part of Meta's expanding hardware plans. According to the report, the company is also working to expand its smart glasses lineup and is considering a subscription service called "Wearables for Work", aimed at business users. The strategy suggests Meta sees wearable technology as a major growth opportunity, particularly as artificial intelligence becomes more integrated into daily life. The company has invested billions of dollars in hardware development through its Reality Labs division, which oversees products related to virtual reality, augmented reality and wearable technology.<h2>Can AI wearables help Reality Labs?</h2> The success of future wearable products could be important for Meta's hardware business. Reality Labs has continued to face financial pressure despite ongoing investment. According to company disclosures, the division reported a loss of around $4 billion during the first quarter of the year. Meta has argued that these investments are necessary to build the next generation of computing platforms, including AI-powered devices and augmented reality products. Analysts say wearable devices could eventually become a key revenue source if consumers embrace AI assistants that work seamlessly throughout the day.

Meta is reportedly working on a new artificial intelligence-powered wearable device that could take the form of a pendant worn around the neck. 


Meta is reportedly working on a new artificial intelligence-powered wearable device that could take the form of a pendant worn around the neck. According to a report by The Information, the company plans to begin testing the device within the next year as it continues expanding its AI hardware strategy. The reported project comes as major technology companies increasingly explore AI-powered devices beyond smartphones and computers. Meta has already invested heavily in smart glasses and virtual reality products, and the new pendant could become another part of its growing wearable ecosystem. What is Meta's AI pendant? According to reports, Meta's new device would function as an AI-powered pendant that users could wear around their neck or attach to clothing. While full details have not been revealed, the wearable is expected to build on technology developed by Limitless, an AI startup acquired by Meta in late 2025. Before the acquisition, Limitless developed a wearable pendant capable of recording conversations, creating summaries and helping users recall information from meetings and daily interactions. Following the acquisition, Meta said the deal would help accelerate its efforts to develop AI-enabled wearable technology. Industry observers believe Meta could integrate its AI assistant into the device, allowing users to interact with artificial intelligence throughout the day without needing to use a smartphone. Why is Meta investing in AI wearables? The move reflects a broader trend across the technology industry. Companies are increasingly exploring devices that allow users to access AI more naturally throughout the day. Instead of opening an app or typing a question, wearable devices can provide hands-free access to AI assistants. Meta CEO Mark Zuckerberg has repeatedly argued that wearable devices could become one of the most important ways people interact with artificial intelligence in the future. The company has already found success with its smart glasses developed in partnership with EssilorLuxottica, which combine cameras, audio features and AI-powered assistance. An AI pendant would represent another step towards making AI available throughout the day in a less intrusive form. Why have AI wearables struggled in the past? Despite growing interest in artificial intelligence, many dedicated AI wearable devices have struggled to gain widespread adoption. Several products launched over the past few years faced challenges related to privacy concerns, battery life, limited functionality and unclear consumer demand. Some users questioned whether wearable devices that continuously listen to conversations or collect information could create privacy risks. Others found that early AI devices did not offer enough practical benefits compared with smartphones. These challenges have led some technology companies to rethink how AI hardware should be designed and marketed. Meta appears to be betting that advances in AI models and wearable technology could make a new generation of devices more useful. Meta's broader wearable strategy The reported pendant project is only one part of Meta's expanding hardware plans. According to the report, the company is also working to expand its smart glasses lineup and is considering a subscription service called "Wearables for Work", aimed at business users. The strategy suggests Meta sees wearable technology as a major growth opportunity, particularly as artificial intelligence becomes more integrated into daily life. The company has invested billions of dollars in hardware development through its Reality Labs division, which oversees products related to virtual reality, augmented reality and wearable technology. Can AI wearables help Reality Labs? The success of future wearable products could be important for Meta's hardware business. Reality Labs has continued to face financial pressure despite ongoing investment. According to company disclosures, the division reported a loss of around $4 billion during the first quarter of the year. Meta has argued that these investments are necessary to build the next generation of computing platforms, including AI-powered devices and augmented reality products. Analysts say wearable devices could eventually become a key revenue source if consumers embrace AI assistants that work seamlessly throughout the day.

Meta is reportedly working on a new artificial intelligence-powered wearable device that could take the form of a pendant worn around the neck.&nbsp;

After AI glasses, Meta is reportedly developing an AI pendant: What it is and how it could work

#Gravitas Meta has turned its AI dream into a global warning about the future of jobs. The company has cut around 8,000 jobs as it pushes aggressively into artificial intelligence.


Meta's AI Push Costs 8,000 Jobs

Leaked audio by labour-focused media organisation More Perfect Union from an April 30 town hall reveals Meta CEO Mark Zuckerberg saying that its employees were used to train AI data. In the audio it Zuckerberg could be hard explaining that the company’s new artificial intelligence data tracking program was a way to “learn from watching really smart people do things.” The audio surfaced as Meta has begun laying off roughly 8,000 employees, 10 per cent of its workforce. In the recording, Zuckerberg defended a mandatory internal software program that logs employee keystrokes, mouse movements, and screen snapshots.Zuckerberg explicitly told employees that the average intelligence of the Meta employees was “significantly higher” than that of the outside contractors typically hired for data labelling. “So if we’re trying to teach the models coding, for example, then having people internally build tools or solve tasks that help teach the model how to code, we think is going to dramatically increase our model’s coding ability faster than what others in the industry have the capability to do, who don’t have thousands and thousands of extremely strong engineers at their company," said Zuckerberg. Meta did not spell out or clarified on the whole process because “this stuff is tricky” &nbsp;and “not strategically in your interest for us to communicate everything” so that competitors could not catch on.Also Read: <a href="https://www.wionews.com/world/nvidia-q1-fy2027-record-revenue-ai-boom-data-centre-growth-1779368905223">NVIDIA posts record $81.6 billion revenue: Why NVIDIA is the biggest winner of the global AI boom</a>

<blockquote class="twitter-tweet" data-media-max-width="560">LEAKED AUDIO: In an all-hands meeting on April 30, Mark Zuckerberg tells employees that he&#39;s training AI on them ahead of mass layoffs. &quot;The AI models learn from watching really smart people do things... The average intelligence of the people who are at this company is… <a href="https://t.co/lt9eeJ3cwh">pic.twitter.com/lt9eeJ3cwh</a>&mdash; More Perfect Union (@MorePerfectUS) <a href="https://twitter.com/MorePerfectUS/status/2056842597117636890?ref_src=twsrc%5Etfw">May 19, 2026</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

Meta has laid off 8,000 active positions and cancelled 6,000 open job roles in European and US offices, as the company is redirecting between $125 billion and $145 billion into AI infrastructure and data centres this year alone. Meta is further assigning 7,000 workers to four new organisations focused on developing AI tools. They will be moved to a new segment of the “AI native design structures” where the employee-to-manager ratio will be smaller than in other organisations.&nbsp;The NYT earlier reported that employees were sceptical about the AI tracking program and expressed fear that they would unknowingly train models that would replace them. The tracking program known as the Model Capability Initiative runs across apps like Gmail, VSCode, and Meta’s internal assistant, Metamate. When employees asked how to opt out, Meta Chief Technology Officer Andrew Bosworth responded bluntly, "There is no option to opt out of this on your work provided laptop."

Leaked audio reveals Mark Zuckerberg saying Meta employees were used to train AI models through internal tracking software, as the company lays off 8,000 workers and ramps up AI investments.

Leaked audio by labour-focused media organisation More Perfect Union from an April 30 town hall reveals Meta CEO Mark Zuckerberg saying that its employees were used to train AI data. In the audio it Zuckerberg could be hard explaining that the company’s new artificial intelligence data tracking program was a way to “learn from watching really smart people do things.” The audio surfaced as Meta has begun laying off roughly 8,000 employees, 10 per cent of its workforce. In the recording, Zuckerberg defended a mandatory internal software program that logs employee keystrokes, mouse movements, and screen snapshots. Zuckerberg explicitly told employees that the average intelligence of the Meta employees was “significantly higher” than that of the outside contractors typically hired for data labelling. “So if we’re trying to teach the models coding, for example, then having people internally build tools or solve tasks that help teach the model how to code, we think is going to dramatically increase our model’s coding ability faster than what others in the industry have the capability to do, who don’t have thousands and thousands of extremely strong engineers at their company," said Zuckerberg. Meta did not spell out or clarified on the whole process because “this stuff is tricky” and “not strategically in your interest for us to communicate everything” so that competitors could not catch on. Also Read: NVIDIA posts record $81.6 billion revenue: Why NVIDIA is the biggest winner of the global AI boom

Leaked audio reveals Mark Zuckerberg saying Meta employees were used to train AI models through internal tracking software, as the company lays off 8,000 workers and ramps up AI investments.

Meta CEO Mark Zuckerberg says 'smart' employees used to train AI models in viral audio

Meta is once again facing scrutiny after another round of layoffs highlighted the harsh realities of the global tech industry slowdown. The job cuts reflect growing pressure on major technology companies to reduce costs, restructure operations and focus on AI-driven growth. Employees and analysts say the layoffs expose rising uncertainty in Silicon Valley despite strong investments in artificial intelligence and digital expansion.


Meta Layoff: The Horror Layoffs at Meta Exposes New Tech Reality

Employees at Meta are reportedly protesting against workplace tracking software that some workers claim monitors mouse activity and productivity inside offices. Staff members have allegedly distributed flyers raising concerns about privacy, surveillance and employee trust. The controversy has sparked a broader debate about how far companies should go in tracking worker performance, especially as tech firms increasingly rely on digital monitoring systems in hybrid and office environments.

Meta Employees Revolt Over Workplace Tracking Software And Mouse Monitoring

In a blow to privacy, Meta ended end-to-end encryption on Instagram. At present, this is the default option on platforms like Signal, WhatsApp and Facebook Messenger. Owing to Meta's move, all content of direct messages on Instagram can be accessed by the platform now. 

Instagram Chats May No Longer Be Private After Meta Ends Encryption

Meta has ended end-to-end encryption for direct messages on Instagram from May 8, 2026. This means that messages sent through Instagram DMs are no longer fully private, marking a significant change in how users communicate on the platform.The feature, which was introduced in 2023, ensured that only the sender and receiver could read messages. With its removal, chats may now be accessible under certain conditions, depending on how systems are used to monitor content.&nbsp;<h2>What changes for Instagram users</h2>With end-to-end encryption removed, Instagram users will notice a change in how their messages are handled.Earlier, encrypted chats meant:Messages could not be accessed by third partiesOnly users involved in the chat could read themNow, messages may be:Analysed for safety purposesAccessible through technical systems if requiredMeta had earlier informed users about this update and provided an option to download or back up chats before the change took effect.<h2>Why Meta is making this move</h2>One of the key reasons behind the change is linked to user safety, especially concerns around harmful content.Reports suggest that removing encryption allows platforms to: Detect child sexual abuse material (CSAM) Identify grooming behaviour Flag harassment or illegal contentThe move follows increasing global pressure on tech companies to improve safety measures on messaging platforms.<h2>Legal and regulatory pressure</h2>Governments in several regions, including the United States, United Kingdom and European Union, have been pushing for stronger monitoring of online platforms.Regulations such as:The UK’s Online Safety ActProposed EU Chat Control rules are aimed at ensuring companies can detect harmful content, even within private communications. Earlier this year, a jury in New Mexico ordered Meta to pay $375 million, citing concerns about platform safety and user protection.<h2>Privacy concerns and debate</h2>While the change may help detect harmful activity, it has also raised concerns among privacy advocates.Experts argue that:Removing encryption may reduce user privacyMessages could become more vulnerable to misuseTrust in private communication may be affectedThis creates a balance issue between ensuring safety and protecting personal data.<h2>What happens next</h2>Meta has not confirmed whether similar changes will be applied to other platforms like Facebook Messenger or WhatsApp. For now, the update applies only to Instagram, but it signals a broader shift in how tech companies may handle private messaging in the future. As regulations continue to evolve, platforms may increasingly adjust their systems to meet safety requirements while managing user expectations around privacy.

Meta has ended end-to-end encryption for direct messages on Instagram from May 8, 2026. This means that messages sent through Instagram DMs are no longer fully private, marking a significant change in how users communicate on the platform.

Meta has ended end-to-end encryption for direct messages on Instagram from May 8, 2026. This means that messages sent through Instagram DMs are no longer fully private, marking a significant change in how users communicate on the platform. The feature, which was introduced in 2023, ensured that only the sender and receiver could read messages. With its removal, chats may now be accessible under certain conditions, depending on how systems are used to monitor content. What changes for Instagram users With end-to-end encryption removed, Instagram users will notice a change in how their messages are handled. Earlier, encrypted chats meant: Messages could not be accessed by third parties Only users involved in the chat could read them Now, messages may be: Analysed for safety purposes Accessible through technical systems if required Meta had earlier informed users about this update and provided an option to download or back up chats before the change took effect. Why Meta is making this move One of the key reasons behind the change is linked to user safety, especially concerns around harmful content. Reports suggest that removing encryption allows platforms to: Detect child sexual abuse material (CSAM) Identify grooming behaviour Flag harassment or illegal content The move follows increasing global pressure on tech companies to improve safety measures on messaging platforms. Legal and regulatory pressure Governments in several regions, including the United States, United Kingdom and European Union, have been pushing for stronger monitoring of online platforms. Regulations such as: The UK’s Online Safety Act Proposed EU Chat Control rules are aimed at ensuring companies can detect harmful content, even within private communications. Earlier this year, a jury in New Mexico ordered Meta to pay $375 million, citing concerns about platform safety and user protection. Privacy concerns and debate While the change may help detect harmful activity, it has also raised concerns among privacy advocates. Experts argue that: Removing encryption may reduce user privacy Messages could become more vulnerable to misuse Trust in private communication may be affected This creates a balance issue between ensuring safety and protecting personal data. What happens next Meta has not confirmed whether similar changes will be applied to other platforms like Facebook Messenger or WhatsApp. For now, the update applies only to Instagram, but it signals a broader shift in how tech companies may handle private messaging in the future. As regulations continue to evolve, platforms may increasingly adjust their systems to meet safety requirements while managing user expectations around privacy.

Meta has ended end-to-end encryption for direct messages on Instagram from May 8, 2026. This means that messages sent through Instagram DMs are no longer fully private, marking a significant change in how users communicate on the platform.

‘Your DM chats are no longer private’: Instagram removes end-to-end encryption from messages

Big changes are coming to Instagram DMs after May 8, as Meta reportedly scales back end-to-end encryption. This move raises concerns about how private conversations will be in the future. Users may no longer have the same level of message security they once relied on. Privacy advocates are already warning that this could weaken trust in one of the world’s most popular messaging platforms.


Instagram Ends Private Chat Feature After May 8 

Meta has challenged a recent court verdict that held the company partly liable for social media addiction among users, arguing that the decision misinterprets how online platforms operate. The case centers on claims that design features on social media apps contribute to excessive usage, particularly among younger users. Meta maintains that responsibility for usage patterns is shared and says it will continue to appeal the ruling as debates over tech regulation and user harm intensify globally.


Meta Challenges Court Ruling Over Social Media Addiction Liability

In March, a court verdict in U.S. took tech giants by a storm. It held Meta accountable for designing social media platforms that are harmful to young people. Meta has now challenged this verdict, citing a federal law which exempts online platforms from liability for user-generated content. 

Meta Appeals Against Verdict Holding It Accountable for Social Media Addiction

A US jury has delivered a landmark verdict against Meta, finding the tech giant negligent over its platform design alongside Google in a case that could reshape the future of social media regulation. Meta has been ordered to pay $4.2 million in damages, sparking fresh debate around online safety, algorithm-driven engagement, and corporate accountability. The ruling is being seen as a major legal moment that could influence future lawsuits against Big Tech companies worldwide.


Meta Ordered To Pay $4.2 Million In Damages

Major publishers have taken legal action against Meta, accusing the company of using their content without permission to train its LLaMA AI models. The lawsuit alleges copyright violations and raises serious questions about how tech giants source data for artificial intelligence. The case could have far-reaching implications for the future of AI development, content ownership, and digital rights worldwide.


Meta AI Training Sparks New Copyright Battle

Lawsuit Alleges Copyright Misuse by Meta, Legal Battle Over AI Training Data Intensifies

Social media giant Meta Platforms is under legal pressure as the U.S. state of New Mexico seeks $3.7 billion in damages. The case targets platforms like Instagram and Facebook, alleging serious concerns over user safety and platform accountability.

The lawsuit marks a significant escalation in regulatory scrutiny of big tech, raising broader questions about social media governance, user protection, and corporate responsibility.

Instagram, Facebook Under Legal Scrutiny: New Mexico State Seeks $3.7 BN From Meta

Meta steps into humanoid robotics with ARI acquisition: What it means for AI strategyMeta has expanded its artificial intelligence efforts by acquiring Assured Robot Intelligence (ARI), a startup that develops AI systems for humanoid robots. The company said ARI works on technology that helps robots understand, predict and respond to human behaviour in complex environments. The financial details of the deal have not been disclosed, but the move signals Meta’s growing focus on robotics alongside its existing AI investments.&nbsp;<h2>What ARI adds to Meta’s AI push</h2>ARI’s work is focused on improving how robots interact with people and surroundings. Its systems are designed to help machines process human actions and adapt in real time. As part of the deal, ARI’s team, including researchers Xiaolong Wang, Lerrel Pinto and Xuxin Cheng, will join Meta’s AI division. Their experience in robotics and machine learning is expected to support Meta’s long-term plans in building intelligent systems that can operate in real-world settings.&nbsp;<h2>Meta’s broader plan for robotics</h2>The newly acquired team will work under Meta Superintelligence Labs, led by Alexandr Wang, and collaborate with Meta’s Robotics Studio. This division is focused on building the core technology required for humanoid robots, including software systems, sensors and AI models. Reports suggest that Meta has been exploring both software and hardware for robots, indicating that the company may eventually develop products aimed at everyday use.&nbsp;<h2>A possible platform for humanoid robots</h2>Industry reports indicate that Meta may be working on a broader ecosystem for robotics. The idea is to build a common software platform that other manufacturers can use, similar to how Android powers smartphones. Meta’s leadership has previously suggested that such technology could be licensed, allowing different companies to build robots using a shared system. This approach could help standardise how humanoid robots operate across the industry.<h2>Competition in the growing robot market</h2>Meta is entering a space where several major companies are already active. Tesla is developing its Optimus robot, while Google is working on AI models designed for robotics. At the same time, Chinese companies are also advancing in humanoid development, with some robots showing strong capabilities in movement and speed. This growing competition highlights the importance of robotics as the next area of technological development.<h2>Shift in focus from metaverse to AI</h2>Meta’s move into robotics also reflects a shift in its strategy. The company had earlier focused on the metaverse but has since reduced updates to its Horizon Worlds platform. It is now investing more in AI technologies, including large language models, AI agents and robotics. This shift suggests that Meta sees AI as a key area for future growth.<h2>What this means for the future</h2>Meta’s entry into humanoid robotics shows how AI is moving beyond software into physical systems. While the company has not announced any products yet, its investments in talent and technology indicate long-term plans. As development continues, the focus will likely remain on building robots that can understand human behaviour, perform tasks and assist in daily activities. The coming years will determine how these efforts translate into real-world use and how Meta competes with other players in this space.&nbsp;&nbsp;

Meta has acquired Assured Robot Intelligence (ARI), a startup focused on AI systems for humanoid robots. The move signals Meta’s push to expand beyond software AI into physical, real-world intelligence systems. ARI’s team will join Meta’s AI division to help build robots.

Meta steps into humanoid robotics with ARI acquisition: What it means for AI strategy Meta has expanded its artificial intelligence efforts by acquiring Assured Robot Intelligence (ARI), a startup that develops AI systems for humanoid robots. The company said ARI works on technology that helps robots understand, predict and respond to human behaviour in complex environments. The financial details of the deal have not been disclosed, but the move signals Meta’s growing focus on robotics alongside its existing AI investments. What ARI adds to Meta’s AI push ARI’s work is focused on improving how robots interact with people and surroundings. Its systems are designed to help machines process human actions and adapt in real time. As part of the deal, ARI’s team, including researchers Xiaolong Wang, Lerrel Pinto and Xuxin Cheng, will join Meta’s AI division. Their experience in robotics and machine learning is expected to support Meta’s long-term plans in building intelligent systems that can operate in real-world settings. Meta’s broader plan for robotics The newly acquired team will work under Meta Superintelligence Labs, led by Alexandr Wang, and collaborate with Meta’s Robotics Studio. This division is focused on building the core technology required for humanoid robots, including software systems, sensors and AI models. Reports suggest that Meta has been exploring both software and hardware for robots, indicating that the company may eventually develop products aimed at everyday use. A possible platform for humanoid robots Industry reports indicate that Meta may be working on a broader ecosystem for robotics. The idea is to build a common software platform that other manufacturers can use, similar to how Android powers smartphones. Meta’s leadership has previously suggested that such technology could be licensed, allowing different companies to build robots using a shared system. This approach could help standardise how humanoid robots operate across the industry. Competition in the growing robot market Meta is entering a space where several major companies are already active. Tesla is developing its Optimus robot, while Google is working on AI models designed for robotics. At the same time, Chinese companies are also advancing in humanoid development, with some robots showing strong capabilities in movement and speed. This growing competition highlights the importance of robotics as the next area of technological development. Shift in focus from metaverse to AI Meta’s move into robotics also reflects a shift in its strategy. The company had earlier focused on the metaverse but has since reduced updates to its Horizon Worlds platform. It is now investing more in AI technologies, including large language models, AI agents and robotics. This shift suggests that Meta sees AI as a key area for future growth. What this means for the future Meta’s entry into humanoid robotics shows how AI is moving beyond software into physical systems. While the company has not announced any products yet, its investments in talent and technology indicate long-term plans. As development continues, the focus will likely remain on building robots that can understand human behaviour, perform tasks and assist in daily activities. The coming years will determine how these efforts translate into real-world use and how Meta competes with other players in this space.

Meta has acquired Assured Robot Intelligence (ARI), a startup focused on AI systems for humanoid robots. The move signals Meta’s push to expand beyond software AI into physical, real-world intelligence systems. ARI’s team will join Meta’s AI division to help build robots.

‘Meta’s robotics plan’: What the ARI deal reveals about the future

The European Commission has formally charged Meta under the Digital Services Act for failing to protect children on its platforms.

Regulators found that age-verification systems on Instagram and Facebook are ineffective, allowing users under 13 to bypass restrictions simply by entering false birth dates. Reporting tools were also flagged as overly complex, sometimes requiring multiple steps to flag underage users.

Meta could now face fines of up to 6% of its global turnover, marking one of the most significant enforcement actions under the EU’s sweeping tech rules.

EU Charges Meta Over Child Safety Failures: Faces 6% Global Revenue Fine

A major tech deal is facing fresh uncertainty as China urges Meta Platforms to unwind its $2 billion agreement with Manus. The move comes amid tightening scrutiny over AI investments and growing US-China tensions in the tech space. With Manus reportedly shifting its headquarters to Singapore, the situation reflects how geopolitical pressures are increasingly shaping the future of artificial intelligence deals.


China Urges Meta to Unwind Manus AI Deal

Meta will install tracking software on the computers of its staff to track mouse movements and keystrokes, and the data will be used to train AI agents. A Reuters report states that the company sent out a memo regarding the action, which is being seen as an open acknowledgement of workers training their own replacements. The report states that the company is ramping up employee monitoring and will use Model Capability Initiative, a surveillance tool and will record the screens of the staff in real time. Further, the company is intensifying internal data harvesting efforts through its "AI for Work" program, which has been rebranded with the corporate-heavy title: Agent Transformation Accelerator. This data will likely be fed to the AI models intended to "accelerate" or automate these roles and improve the functionality of its agents.Also Read: <a href="https://www.wionews.com/technology/google-ends-back-button-hijacking-will-de-rank-websites-for-phantom-pages-1776237638909">Google ends back button hijacking, will de-rank websites for 'phantom pages': Is yours at risk?</a><h2>What Meta told its employees</h2>The aim is to have them operate autonomously and navigate across multiple systems and programs. According to Reuters, Meta’s current agent offerings are unable to perform certain actions that humans can easily do, such as selecting items from dropdown menus or using keyboard shortcuts. By tracking mouse movements and keystrokes, Meta is vying to refine those capabilities in its agents. To ease employee concerns, Meta has told them that their work won't be affected because of the continuous monitoring. The intrusion has been termed as an opportunity to "help our models get better simply by doing their daily work." They have also been promised that the data would not be used for performance reviews. However, the move has created fear since reports suggest that Meta is planning to lay off 10 per cent of its staff in the coming months.<h2>Mera investing in agents, &nbsp;Zuckerberg chatbot an example</h2>Besides, the employees have taken on an additional role of training AI while sacrificing their own privacy for no additional compensation. They are clearly training AI agents to do their jobs as Meta has been investing heavily in the field. They are reportedly building AI agents that can work alongside humans. The company even has a Zuckerberg chatbot for employees to communicate with.&nbsp;<h2> Netizens react to surveillance ploy</h2> People reacted to the news as expected. A user asked, "So they're basically telling the employees to train AI models that would replace them? Am I think correctly." Another wrote, "Unreasonable search and seizures comes to mind. This seems like a clear constitutional violation." One user asked, "If productivity data becomes training data, where do we draw the line between optimization and employee surveillance?" One suggested to the staff at Meta, "They should all collectively walk out. &nbsp;Seriously why the hell do this." &nbsp;

Meta will monitor the screens, keystrokes and mouse movements of its workers in real time and use the data to train AI agents. The employees have been told not to worry, as they are helping the models get better. Are they training their replacements?

Meta will install tracking software on the computers of its staff to track mouse movements and keystrokes, and the data will be used to train AI agents. A Reuters report states that the company sent out a memo regarding the action, which is being seen as an open acknowledgement of workers training their own replacements. The report states that the company is ramping up employee monitoring and will use Model Capability Initiative, a surveillance tool and will record the screens of the staff in real time. Further, the company is intensifying internal data harvesting efforts through its "AI for Work" program, which has been rebranded with the corporate-heavy title: Agent Transformation Accelerator. This data will likely be fed to the AI models intended to "accelerate" or automate these roles and improve the functionality of its agents. Also Read: Google ends back button hijacking, will de-rank websites for 'phantom pages': Is yours at risk? What Meta told its employees The aim is to have them operate autonomously and navigate across multiple systems and programs. According to Reuters, Meta’s current agent offerings are unable to perform certain actions that humans can easily do, such as selecting items from dropdown menus or using keyboard shortcuts. By tracking mouse movements and keystrokes, Meta is vying to refine those capabilities in its agents. To ease employee concerns, Meta has told them that their work won't be affected because of the continuous monitoring. The intrusion has been termed as an opportunity to "help our models get better simply by doing their daily work." They have also been promised that the data would not be used for performance reviews. However, the move has created fear since reports suggest that Meta is planning to lay off 10 per cent of its staff in the coming months. Mera investing in agents, Zuckerberg chatbot an example Besides, the employees have taken on an additional role of training AI while sacrificing their own privacy for no additional compensation. They are clearly training AI agents to do their jobs as Meta has been investing heavily in the field. They are reportedly building AI agents that can work alongside humans. The company even has a Zuckerberg chatbot for employees to communicate with. Netizens react to surveillance ploy People reacted to the news as expected. A user asked, "So they're basically telling the employees to train AI models that would replace them? Am I think correctly." Another wrote, "Unreasonable search and seizures comes to mind. This seems like a clear constitutional violation." One user asked, "If productivity data becomes training data, where do we draw the line between optimization and employee surveillance?" One suggested to the staff at Meta, "They should all collectively walk out. Seriously why the hell do this."

Meta will monitor the screens, keystrokes and mouse movements of its workers in real time and use the data to train AI agents. The employees have been told not to worry, as they are helping the models get better. Are they training their replacements?

Meta to spy on staff's clicks and keystrokes to train AI agents, netizens ask 'training AI models that would replace them?'

Meta is reportedly developing an advanced AI version of its CEO Mark Zuckerberg as part of its broader push into artificial intelligence and digital assistants. The AI model is expected to simulate Zuckerberg’s communication style, decision-making patterns, and public interactions, potentially being used for internal tools, user engagement, and product demonstrations across Meta’s platforms.


Meta Developing AI Version of its CEO Mark Zuckerberg

AI companies are launching new models at a rapid pace. In the latest development, Meta has introduced a new AI model called Muse Spark, which the company says is its most advanced system so far. The model was developed by Meta Superintelligence Labs (MSL), a new research division set up by Meta in 2025. The team is led by Alexandr Wang, who joined the company after Meta acquired his startup Scale AI in a multi-billion-dollar deal.According to company announcements and developer discussions, Muse Spark is designed to power Meta’s AI assistant across platforms such as Facebook, Instagram, and WhatsApp.

<blockquote class="twitter-tweet">1/ today we&#39;re releasing muse spark, the first model from MSL. nine months ago we rebuilt our ai stack from scratch. new infrastructure, new architecture, new data pipelines. muse spark is the result of that work, and now it powers meta ai. 🧵 <a href="https://t.co/fThDXdsxwB">pic.twitter.com/fThDXdsxwB</a>&mdash; Alexandr Wang (@alexandr_wang) <a href="https://twitter.com/alexandr_wang/status/2041909376508985381?ref_src=twsrc%5Etfw">April 8, 2026</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

<h2>What is Meta Muse Spark?</h2>Muse Spark is the first model from Meta’s new Muse family of AI systems, which is expected to replace the company’s earlier Llama models over time. Unlike some large flagship models, Muse Spark is described as a small but fast AI model. Meta says the model was built from scratch in about nine months.Despite its smaller size, the company claims the model can still handle complex reasoning tasks. In its technical description, Meta said the system can process questions related to science, mathematics and health.According to Alexandr Wang, the model also includes a “contemplating mode”, where multiple AI agents can work in parallel to solve complex problems.<h2>How Muse Spark compares with ChatGPT and Claude</h2>Muse Spark enters a highly competitive market where several companies are developing advanced AI assistants.Major competitors include:ChatGPT developed by OpenAIClaude developed by AnthropicMeta claims that Muse Spark performs competitively in certain benchmarks, including:Multimodal reasoningScientific problem solvingHealth-related analysisAgent-based workflowsThe company also said early testing showed the model could compete with other advanced reasoning systems such as Google’s Deep Think models and OpenAI’s professional reasoning models.However, these comparisons are based on internal tests and have not yet been widely verified by independent researchers.<h2>What can Muse Spark do?</h2>Meta says the new AI system is designed to understand real-world context and assist users across its ecosystem of apps.Some of its capabilities include:Solving science and mathematics problemsAnswering complex questionsManaging multiple AI agents for tasksProviding recommendations for shopping and creatorsUnderstanding conversations across Meta appsThe model will power an updated version of Meta AI, which the company plans to integrate across Facebook, Instagram, WhatsApp, Messenger and Meta’s smart glasses.Meta says the system is being built for the three billion people who use its platforms globally.Muse Spark is not open source. One major difference from Meta’s earlier AI models is that Muse Spark is currently closed source.Meta’s previous Llama models were released publicly, allowing developers to access the code and build applications. Muse Spark, however, is currently available only through a private API preview for selected partners. The company said future versions of the Muse models may be open sourced, but no timeline has been confirmed.<h2>Safety testing and future development</h2>Meta said the new AI model has undergone safety evaluations, particularly in areas such as biological or chemical misuse.According to Alexandr Wang, Muse Spark showed strong refusal behaviour when asked about high-risk topics during internal testing.The company added that larger AI models in the Muse family are already under development, with new infrastructure being built to support them.This suggests Muse Spark may be only the first step in Meta’s long-term AI roadmap.<h2>Why Muse Spark matters in the AI race</h2>The launch of Muse Spark reflects the growing competition among technology companies to build powerful AI assistants. Companies including Meta, OpenAI, Anthropic and Google are investing billions of dollars into AI research and infrastructure. Meta’s strategy is different in one key way: the company plans to integrate AI deeply across its social platforms, where billions of users interact every day.If Muse Spark performs as expected, it could play a major role in shaping how people use AI inside messaging apps, social media and digital services.

Meta has introduced Muse Spark, the first model in its new Muse AI family developed by Meta Superintelligence Labs. Led by Alexandr Wang, the model is designed to power Meta AI assistants across Facebook, Instagram and WhatsApp.

AI companies are launching new models at a rapid pace. In the latest development, Meta has introduced a new AI model called Muse Spark, which the company says is its most advanced system so far. The model was developed by Meta Superintelligence Labs (MSL), a new research division set up by Meta in 2025. The team is led by Alexandr Wang, who joined the company after Meta acquired his startup Scale AI in a multi-billion-dollar deal. According to company announcements and developer discussions, Muse Spark is designed to power Meta’s AI assistant across platforms such as Facebook, Instagram, and WhatsApp.

Meta has introduced Muse Spark, the first model in its new Muse AI family developed by Meta Superintelligence Labs. Led by Alexandr Wang, the model is designed to power Meta AI assistants across Facebook, Instagram and WhatsApp.

What is Muse Spark? Meta’s newly launched AI model and how it compares to ChatGPT and Claude

Meta Platforms’s reported acquisition deal for AI startup Manus AI is valued at around $2 billion, highlighting the tech giant’s aggressive push to strengthen its position in the global artificial intelligence race. The Singapore-based firm, known for developing advanced AI agents capable of performing complex tasks autonomously, is expected to play a key role in expanding Meta’s AI ecosystem across consumer and enterprise platforms.


China Reviews Meta's $2 Billion Manus AI Deal 

Social media platforms are built to capture attention, using powerful algorithms to keep users scrolling longer. Experts warn this “attention economy” model may be contributing to rising mental health concerns, including anxiety and low self-esteem. Studies suggest the constant need for engagement can make it harder to disconnect, raising fresh questions about how digital platforms shape behaviour and well-being.


meta