OpenAI releases powerful cybersecurity model to general public
OpenAI has publicly released advanced artificial intelligence models with broad cybersecurity features, creating new technology and risk management challenges for financial institutions.
FTMQ Business, written by our newsroom1 view

OpenAI put artificial intelligence models with broad cybersecurity features on open sale this week, PYMNTS reported. The release comes two weeks after OpenAI restricted its most capable cybersecurity model to vetted security teams due to high risk concerns. PYMNTS reported that the GPT-6.2 model carries the same cyber capabilities that were previously restricted, forcing banks to assume attackers possess the same tools as their defenders. [1]
The release increases competition among major technology companies targeting corporate software functions. According to PYMNTS, OpenAI claims its GPT-6.2 Astra beats Anthropic’s Claude Fable 5.2 on legal tasks at one-sixth of the cost and leads on finance deliverables including valuation workbooks. Meanwhile, Google’s Gemini 4 Neon ranks first on legal and tax benchmarks at less than half the price of frontier models. [1]
The tools rely on reasoning language models, which are large language models trained to complete tasks requiring multiple steps of logical thinking. These systems outperform standard language models on logic, programming, and mathematical calculations by revising earlier steps and deploying extra computations. [2]
The operational shift follows previous infrastructure challenges for major technology providers in the financial sector. As FTMQ Business reported earlier, a power failure at a Google Cloud data centre in London disrupted major institutions including Revolut, Halifax, HSBC, and Lloyds in October 2026. [5]
In short
- OpenAI released its GPT-6.2 model with cybersecurity capabilities to the open market.
- OpenAI restricted its top cybersecurity model to vetted security teams two weeks prior to the public release.
- OpenAI claims GPT-6.2 Astra performs legal tasks at one-sixth the cost of Anthropic's Claude Fable 5.2.
- Google's Gemini 4 Neon model leads legal and tax benchmarks at under half the frontier model price.
Sources
Every paragraph above points to the numbered items it rests on. Read the originals here.
- [1]OpenAI Unlocks Cyber Power as Google Slashes AI CostsPYMNTS, 1d ago (the report this story comes from)
Background
- [2]Reasoning model on Wikipedia
- [3]Qwen on Wikipedia
- [4]Fervo Energy shares surge following Cape Station launch and Google deal FTMQ Business, 11d ago
- [5]Google Cloud Power Failure Disrupts Major United Kingdom Banks and Services FTMQ Business, 31d ago
Our newsroom writes these reports with the help of software, from the 5 sources listed and nothing else, and checks them against those sources. Facts can still be wrong or move on; the originals are the record. Spotted a mistake? Write to daniel@monsterkong.com.
Related from FTMQ Business
Earlier reports of ours on the same people and subjects.
- OpenAI cuts internet access during advanced artificial intelligence model trainingBanking & Credit, 26d ago
- Google Cloud Power Failure Disrupts Major United Kingdom Banks and ServicesEconomy, 31d ago
- US banking regulators extend comment deadline for third party risk guidelinesEconomy, 7d ago
- Think tank calls for independent reviews of high risk Bangladeshi banksBanking & Credit, 6d ago
- Reflection AI Launches Open Model Beam to Compete With Chinese RivalsTop Business, 46d ago
More in Banking & Credit
- XRP Ledger lending protocol awaits approval as token price remains depressed15h ago
- British Digital Bank OakNorth Secures United States National Banking Charter15h ago
- Citi warns customers using third-party artificial intelligence agents on its platforms15h ago
- PayPal uses artificial intelligence agent to track $53 billion in cash4d ago
- Federal Reserve Uses Scenario Models to Assess Artificial Intelligence Impacts4d ago
- Afterpay Co-Founder Nick Molnar to Leave Block at Year End5d ago
Get the day in one email
Reports like this one, the top news of the last 24 hours, every morning. Free, one email a day; readers can comment under every report.
By signing up you agree to our terms and privacy policy. Unsubscribe any time.

Comments
Loading
Join the conversation
Comments are open to readers of our daily email: the top news of the last 24 hours, every morning, free. Sign up and the comment box opens.
Already on the list? Enter the same address and we will send a sign-in link.
By signing up you agree to our terms and privacy policy. Unsubscribe any time.