&imwidth=800&imheight=600&format=webp&quality=medium "OpenAI will pay up to $50,000 if you can do this")
OpenAI is asking security researchers to try to break one of its most advanced AI models and it is willing to pay up to $50,000 for it.
The company has expanded its Bio Bug Bounty Programme to include GPT-5.6, doubling the top reward from the previous $25,000 offered during the GPT-5.5 programme. The initiative focuses on finding "universal jailbreaks" that can consistently bypass the model's biology-related safety safeguards. The move comes as AI companies face growing scrutiny over how advanced models could be misused to generate harmful biological information. OpenAI says discovering vulnerabilities before attackers do is an important part of making future AI systems safer.
What is OpenAI looking for?
Unlike ordinary prompt tricks that may work only once, OpenAI wants researchers to discover universal jailbreaks techniques that reliably bypass the model's built-in biosafety protections across multiple conversations. According to the company, these findings will help improve future safety systems while reducing the chances of malicious misuse. Researchers who discover smaller weaknesses may also receive partial rewards, depending on the quality and impact of their submissions.
Why has the reward increased?
OpenAI says GPT-5.6 represents its latest generation of AI models, making safety testing even more important. To encourage more researchers to participate, the company has doubled the maximum reward to $50,000. The updated programme will replace GPT-5.5 after 27 July 2026, when testing for the older model officially ends. The company added that researchers already enrolled in the GPT-5.5 programme do not need to submit another application.
Who can apply?
Anyone interested in participating must apply through OpenAI's Bio Bug Bounty page. Applicants are required to provide their name, organisation or affiliation, and relevant research experience. Participants must also have a ChatGPT account and sign a non-disclosure agreement (NDA) before gaining access to the testing platform. The GPT-5.6 programme will become available after the GPT-5.5 testing period concludes later this month.
Trending Stories
&im=FitAndFill=(700,400))
How fast can the Sukhoi Su-30 fighter jet really fly?
)
Did China ask its people 'not to respond to aliens'? Origin of bizarre claim on social media
)
MS Dhoni-CSK Fallout? Report claims failed stake sale talks triggered rift
)
Scientists reveal the SAFEST COUNTRY when catastrophe strikes the planet
)
Serial killer on the loose in South Korea's Jeju Island? Fourth body found in three days as more than a dozen people go missing
)
2/3rd of glacier detached, collapsed: ISRO satellite images reveal what triggered Nepal floods
)
Trump once again claims without him there would be ‘no Israel right now’
)
Foreign espionage attacks against German companies surge to 37%, Bitkom report reveals
)
Blake Lively wins partial legal fee from Justin Baldoni after $8M request
)
Why is Pakistan deploying Turkish Bayraktar TB2 drones near India border?
)
'OH MY GOD! We've found other agents!': Investigators discover not one or two, almost 700 OpenAI AI agents went rogue, hacked Hugging Face
)
Lunar eclipse 2026: India timings, places to see and more. August Blood moon factsheet
)
RSS chief Mohan Bhagwat on global outreach tour
)
Both pilots killed as training aircraft crashes in Lucknow – What we know
)
16-year-old charged with preparing terror attack on kindergarten in Norway
&im=FitAndFill=(700,400))
How the S-400 protects itself from enemy attacks
)
Nepal floods: Over 165 dead, 400 missing; US, India, China, World Bank mobilise aid
)
India's multi-layered watch: The new surveillance war
)
5 doomsday planes spotted in US skies hours after CIA director's secret trip to Russia
)
9/11 mastermind Khalid Sheikh Mohammed to finally face trial in 2028, nearly 27 years after horrific attacks
)
Islamabad hospital fire: What we know about one of the biggest facility that housed Imran Khan
)
Workers run for life while filming Nepal flash flood: Scary new visuals emerge from Rasuwa
)
Russian military planes enter South Korean air defence zone amid rising tensions with North Korea: What's happening
)
Ron Weasley Is Back! Rupert Grint set to return to Harry Potter universe on Broadway
)
400-year-old 'vampire' grave in Poland reveals non-existent genetic lineage
)
Dolly Parton, global icon and queen of country music, dies at 80
)
Nepal flash floods: Kailash Mansarovar yatra affected, close to 400 pilgrims out of contact
)
'If he is dead, they're putting on a pretty good show': Trump casts fresh doubt over Mojtaba Khamenei's fate
)
Man arrested near US Capitol with guillotine in pickup truck: What happened?
&im=FitAndFill=(700,400))
Why the Su-30MKI cockpit is painted bright turquoise green
Bug bounty programmes have long been used in cybersecurity to uncover software vulnerabilities before criminals can exploit them. OpenAI is applying the same approach to artificial intelligence.
As AI systems become more capable, governments and regulators are placing greater emphasis on testing model safeguards, particularly in areas such as biology, cybersecurity and chemical research. By inviting independent researchers to stress-test GPT-5.6, OpenAI hopes to identify weaknesses early and strengthen its safety systems before the technology reaches wider use.
