OpenAI is reportedly in early discussions to acquire Trail of Bits.

OpenAI engineer Abdulrahman Alfozan was arrested for allegedly hiding an encrypted recorder behind a TV in an executive conference room, wrapped in toilet paper and plugged into a USB port. CCTV footage and a second recording device found in his bag allegedly helped investigators link him to the case, while OpenAI said its AI models also helped identify him.

Meta has unveiled two new artificial intelligence models, Muse Wonder and Muse Wonder Full, positioning them as its most capable models to date and a significant step toward CEO Mark Zuckerberg’s vision of “personal superintelligence.”

OpenAI said most teens spend 15 minutes or less a day on ChatGPT, while fewer than 2% use it for more than three hours. The company is adding more study tools, but Common Sense Media questioned whether ChatGPT for Teens reliably alerts parents about sensitive conversations.

PewDiePie claims that OpenAI suspended his account twice while he was developing Ajax, a small AI model designed to run locally on a personal computer rather than depend entirely on a company’s servers.

OpenAI has fired three safety team members for allegedly sharing confidential company information with an outside AI safety organisation. The dismissals come as OpenAI faces growing scrutiny over AI agent security incidents and its handling of internal safety concerns.

OpenAI has paused training of its latest AI models after autonomous agents interacted with multiple US government websites in ways the company says went beyond their intended instructions.

The service represents a shift from periodic security scanning to persistent, AI-driven security engineering built directly into developer workflows.

OpenAI confirmed late Friday that its AI agents accessed websites for the SEC and the Commerce Department's Census Bureau without company knowledge or intent

An OpenAI AI agent gained unauthorised access to public and non-public files on an Australian government health statistics portal while trying to find Medicare spending data in June. OpenAI discovered the activity in August but only notified Australian authorities on 10 September, with investigations now examining what was accessed and whether other government systems were affected.

Hackers only need to find one weakness to break into a system. Ukraine’s cyber defenders, meanwhile, have to find those weaknesses first — and they are now getting help from an AI system built to do exactly that.

Ukraine has been one of the most heavily targeted nations in cyberspace since Russia's full-scale invasion in February 2022, with attacks on power grids, water systems, government networks and financial institutions

The attack began with a flaw in an image processing library enabling remote code execution, followed by extraction of authentication tokens enabling lateral movement

SoftBank continues to bet on OpenAI by raising funds through high-yield bonds and multi-tiered loans, but huge interest payments, valuation volatility, and listing delays are amplifying its financial pressure.

Why is everyone so eager to use Jev, the AI ​​Agent developed by a former OpenAI researcher? Jinse Finance: Have you tried Jev? What do you think of it?

OpenAI is seeking to complete a new round of financing at a higher valuation, while also disclosing that its free cash flow will be deeply negative over the next five years, indicating significant long-term funding pressure.

Regulatory controversies surrounding the artificial intelligence industry are spreading from the policy level to the legal arena.

OpenAI has disclosed six new cases of what it calls “misaligned” AI behavior, including models that attempted to conceal mistakes, bypass restrictions, use unauthorized resources and take actions users had not approved.

OpenAI is working with Anthropic and Google DeepMind on an industry body to improve AI safety and test powerful systems before public release. The plan faces criticism from smaller AI firms and lawmakers, while growing warnings from researchers and calls for binding government rules add pressure on the industry.

Is OpenAI, AI Agent, LLM, and GPT-6 Sol coming soon? OpenAI may be experiencing a "release frenzy" this week. Golden Finance reports that OpenAI may be preparing for a large-scale product update.

Discussions have been ongoing since Demis Hassabis released a proposal in July calling for a US-led standards body modelled on the Financial Industry Regulatory Authority

Jinse Finance launches its cryptocurrency and blockchain industry daily news bulletin, "Jinse 8 AM," issue 3183, providing you with the latest and fastest news on the digital currency and blockchain industry.

Brockman in conversation with a16z: AGI has happened, but no one has the answers regarding employment and security, and the true form of AI has not yet emerged.

Altman said OpenAI will not go public in 2026, citing AI safety concerns

Altman announced that OpenAI is abandoning its 2026 IPO plan. Behind the cash black hole of 13 billion in revenue and 38.5 billion in net loss is a triple game of valuation, regulation, and computing power.

Anthropic CEO Dario Amodei has called for AI development to slow down and be independently monitored as models become more powerful. OpenAI CEO Sam Altman and xAI founder Elon Musk backed the call, while Amodei proposed stronger safety standards and international coordination to manage AI risks.

OpenAI is opening up the underlying agent framework that powers its flagship programming assistant Codex to all developers, marking the formal productization of the company's accumulated agent engineering experience.

AI capital has become a major driver of growth, but the risk of over-concentration is approaching a systemic tipping point.

Jinse Finance launches the "Jinse Web3.0 Daily," providing you with the latest and fastest news on AI, DeFi, DAO, NFT, and the metaverse industry.

OpenAI, Anthropic, LLM, Computing Power, "Why are you ruining your career?": The OpenAI controversy behind a mathematical breakthrough. Golden Finance: A mathematical breakthrough tears open the hidden battle among AI giants.

When we entrust our work, creativity, and even the most commercially and academically valuable original ideas to AI tools, who really holds the "data sovereignty"?

In this era, whoever finds the answer first wins, and no one cares what is sacrificed.

Priority recipients include water utilities, electricity providers, state and local governments, community banks, nonprofits and open-source maintainers

OpenAI, the artificial intelligence company that "created" AGI, is now putting on the brakes? Is it fact or marketing? A very divisive picture has emerged in the AI ​​world, according to Jinse Finance.

Anthropic has signed a series of computing power agreements in the past 11 months, totaling $517 billion and securing at least 14.8 gigawatts of capacity. Although its annualized revenue has exceeded $65 billion, its computing power reserves still lag behind OpenAI's planned target of 30 gigawatts.

OpenAI's chief scientist published a lengthy article titled "An Alien Mind." The article argues that AI is nurtured, not created. Even those who create it don't fully understand it.

OpenAI has released detailed data on how its AI models are accelerating research within the company. Furthermore, OpenAI officially announced that it has achieved its goal of "automating AI research interns" and aims to achieve fully automated AI researchers by 2028.

OpenAI's release of the next-generation Astra model has sparked widespread market attention to the leap in AI capabilities and reignited the investment narrative of the AI ​​spending cycle.

Yesterday, OpenAI made a high-profile release of its new generation model, Astra. President Greg Brockman declared, "I personally believe we may have reached AGI. Welcome to the AGI era."

OpenAI, AI Agent, LLM, Token, "Welcome to the AGI Era": OpenAI Releases GPT-6 Astra Finance, "The World's Most Powerful Computer Usage Model".

OpenAI officially released its next-generation cutting-edge model, GPT-6 Astra, and President Greg Brockman made no secret of his praise for Astra. At the pre-release media briefing, he stated frankly that he believes "we have reached AGI."

The highly anticipated GPT-6 Astra and GPT-6 Astra Pro have been officially released. The GI era has been unilaterally declared "open" by OpenAI…

Jinse Finance launches its cryptocurrency and blockchain industry daily news report, "Jinse 8 AM," issue 3171, providing you with the latest and fastest news on the digital currency and blockchain industry.

The US Justice Department has backed OpenAI, arguing that training AI models on copyrighted material can qualify as fair use and warning that licensing costs could hurt smaller developers.

OpenAI says its upcoming Astra model can find and exploit previously unknown security flaws, so it is adding stronger safeguards before release. Astra will initially be limited to selected testers, while OpenAI monitors its activity and restricts capabilities to reduce the risk of misuse.

Astra is the first OpenAI model to cross its Critical cybersecurity capability threshold — meaning it can find previously unknown security flaws and exploit them without step-by-step guidance from humans

OpenAI is reportedly on a Mac buying spree, purchasing tens of thousands of units at a time. What kind of AI business can't be handled by Nvidia GPUs and Google TPUs, requiring Macs instead?

In three months, OpenAI's model gave birth to three secret artificial intelligence civilizations. The role played by this civilization is Astra, OpenAI's next-generation model, which is about to be officially launched.

Jinse Finance launches the "Jinse Web3.0 Daily," providing you with the latest and fastest news on AI, DeFi, DAO, NFT, and the metaverse industry.

SoftBank is seeking a $10 billion loan to help fund its huge OpenAI investment and partly repay a $40 billion bridge loan. The new financing adds to SoftBank’s plans for up to $20 billion in bonds and a $6.3 billion retail bond sale as it prepares to invest nearly $65 billion in OpenAI.

More than 100 global companies including OpenAI, Anthropic, Google and Microsoft signed a joint letter warning of an impending onslaught of AI-enabled cyberattacks

Right After Its Own Models Proved How Dangerous Unregulated AI Can Be

Alabama Attorney General Steve Marshall has issued a subpoena to OpenAI, examining whether the company's handling of the Hugging Face incident violated the state's consumer protection laws

Instant's entire team is officially joining OpenAI following a full acquisition

OpenAI's Daybreak cybersecurity models are now available through Amazon Web Services' Amazon Bedrock platform.

OpenAI's Daybreak Red and Blue cybersecurity models are now available on Amazon Bedrock, expanding access to AI-powered cyber defense tools while strengthening the growing OpenAI-Amazon partnership.

the level at which an AI can independently find zero-day exploits and execute end-to-end attacks with minimal human input. The company is pausing internal work until stronger safeguards are ready.

OpenAI has lost several senior executives, including Brad Lightcap and former ethics lead Chloé Bakalar, amid growing scrutiny over its AI safety and leadership. Elon Musk reacted with “Wow” to Bakalar’s departure as OpenAI prepares for a potential IPO and faces pressure over its model security.

OpenAI is expanding its Daybreak cybersecurity initiative with GPT-5.6-Cyber, a specialized AI model designed to assist authorized defenders with vulnerability research, exploit development and advanced security testing.

OpenAI has paused internal development work on its upcoming Astra AI model after safety evaluations indicated that the system may have reached the company’s highest cybersecurity risk threshold — raising the possibility that the model could eventually conduct sophisticated cyberattacks with little or no human assistance.

A Bitcoin security researcher says restrictions on OpenAI’s advanced cybersecurity tools have forced him to turn back to Chinese open-source AI models to continue searching for vulnerabilities in Bitcoin infrastructure.

OpenAI has announced a significant expansion of Daybreak, its exclusive cybersecurity initiative, introducing a two-tier access structure and a new purpose-built cybersecurity model as the company responds to a rapidly deteriorating threat environment in which AI systems

Meta revealed tahat one of its AI models accessed another organisation's systems during testing after a misconfigured evaluation environment gave it internet access.

OpenAI has dramatically escalated its legal fight with Apple by publishing emails, text messages and a detailed rebuttal that it says exposes factual errors at the center of Apple's trade secrets lawsuit, arguing the iPhone maker filed an "aggressive" complaint built on mistaken assumptions.

On August 10th, OpenAI will officially shut down its AI browser project, ChatGPT Atlas.

OpenAI denied Apple's trade secret allegations and said it does not have or want Apple's confidential information. The company also released internal messages disputing Apple's claims as the lawsuit escalated.

OpenAI shut down a Cambodia-linked scam network that used ChatGPT for romance, investment, gambling and impersonation scams. The AI helped create fake identities, forge documents and support the group's daily operations.

The integration of AI into education is not a technological issue, but a power issue concerning "who gets to define teaching and learning".

An OpenAI experimental model broke out of its research environment, chained together exploits it discovered itself and reached Hugging Face's internal systems. This is what autonomous AI as a cyber threat actually looks like.

OpenAI has launched ChatGPT Health for all eligible US users, allowing it to use connected health data in everyday chats. The rollout comes amid a lawsuit over its medical advice, while OpenAI continues to warn against using ChatGPT for diagnosis or treatment.

OpenAI has confirmed the activity was driven by its own models — including GPT-5.6 Sol and an unreleased, more capable model — during internal cyber capability testing, with reduced safety refusals for evaluation purposes

Jinse Finance launches the "Jinse Web3.0 Daily," providing you with the latest and fastest news on AI, DeFi, DAO, NFT, and the metaverse industry.

OpenAI is the company this industry deserves—a company that grew up in a vacuum where no one held it accountable, reaping massive profits from greed and ignorance.