‘Stupid’ AI can’t threaten humanity, says Meta executive Nick Clegg

Even as experts argue whether Artificial Intelligence (AI) is a boon or bane for humanity, Meta’s president of global affairs Nick Clegg believes that the current set of technology doesn’t have the capacity to do either.

“The hype has somewhat sort of run ahead of the technology,” Nick Clegg, president of global affairs at Facebook and Instagram parent firm Meta, told the BBC on Wednesday (July 19).

“I think a lot of the warnings of existential threats relate to models that don’t currently exist, so-called superintelligent, super powerful AI models. In other words, the vision of AI where AI develops some autonomy and agency and it can think for itself, it can reproduce itself.”

Add WION as a Preferred Source

Meta comes up with own AI chatbot—Llama 2

The remarks from Clegg, who is also the former UK deputy prime minister, came after Meta announced that it is making its own AI large language model (LLM), Llama 2.

Mark Zuckerberg said that Llama 2 would be open source and available for free to the public.

According to Clegg, Large Language Models is a platform that power chatbots much like ChatGPT.

Meta introduces generative AI model 'CM3leon' for text, images

Llama 2 will use machine learning to answer questions, compose emails or write essays, providing human-like responses.

“It is basically joining dots in enormous datasets of text, and guessing the next word in a sequence,” he said.

With the proposed launch of Llama 2, Meta seeks to heat up the AI bot market, which is currently dominated by Microsoft-backed OpenAI’s ChatGPT.

Rival to ChatGPT

ChatGPT created quite a buzz and took the tech world by storm last year and sort of catapulted the demand to integrate AI in the business.

Clegg believes that existential threat warnings issued by some AI experts relate to systems which don't yet exist.

But the tech firm’s decision to provide access to its AI to all has stoked concerns about its potential misuse by cybercriminals.

However, Meta insists that its AI cannot be used to promote violence, build computer viruses, for building weapons orfor nuclear technology. It also bars spreading spam, hate speech, or child abuse content.

Nick asserted that Meta’s AI model was “safer” than rivals and had been fined tuned by hundreds of attempts to break it.

When asked about the risks the models posed, Nick said they “cannot build a nuclear bomb”.

“It’s not as if we’re at a T-junction where firms can choose to open source or not. Models are being open-sourced all the time already,” he added.

(With inputs from agencies)

WATCH WION LIVE HERE