OpenAI expects future AI models to be capable of identifying and exploiting vulnerabilities ...

OpenAI expects future AI models to be capable of identifying and exploiting vulnerabilities that are currently unknown, according to a blog post.

Statements of this kind from frontier AI developers are capability signalling rather than disclosure: OpenAI has repeatedly framed upcoming models as reaching thresholds in cyber offense, and each prior instance has served a dual audience, customers and enterprise buyers on one side, regulators and safety frameworks on the other. The pattern in such announcements is that language zero-day discovery tends to precede two follow-ons: tiered access controls or staged releases for the model in question, and a fresh round of policy engagement in Washington, since offensive cyber capability is one of the few AI risk claims with an established regulatory vocabulary. For the listed names in the supply chain, the transmission channel is indirect: it is not a revenue event, but past episodes of heightened capability rhetoric have fed into the AI capex and demand narrative that currently anchors the cohort's multiples, while also raising the question of compute and safety spend at the developer itself. Worth observing is whether the post is tied to a specific model release cycle or is standalone positioning, and whether it elicits any response from national security or cyber agencies, which has been the usual step when labs publicly claim offensive capability. As a blog post rather than a filing or product launch, the signal is trajectory and regulatory posture, not near-term fundamentals.

23 Oct, 18:24 Texas Attorney General Ken Paxton sues OpenAI and CEO Sam Altman for lying to users and harming Texans

23 Oct, 16:05 OpenAI reveals plans for an in-house exabyte-scale storage system

23 Oct, 15:19 CrowdStrike (CRWD) CEO says in the coming months we will see a lot of breaches from the PHP vulnerability

23 Oct, 11:55 Oracle (ORCL) to invest USD 1bln in partner ecosystem over 4 years

23 Oct, 05:29 Newsquawk Daily European Opening News - 23rd October 2026

Newsquawk Weekly Economic Release 26th - 30th October 2026 5 hours ago

US EQUITY OPEN: Breadth poor at the open: Energy leads gains on oil rally, Tech sees slight bounce 5 hours ago

Daily US Earnings Estimates: 26th October 2026: VZ, WELL 1 hour ago

The Trump administration plans to introduce voluntary “US Caught Seafood” labels to help consumers identify seafood origins, according to CBS 3 hours ago

Republican Senator Blackburn has raised concerns Meta’s (META) plans to update its commercial terms, warning that the changes could “exploit American creators and businesses,” according to a letter 3 hours ago

CNBC speculates that Verizon (VZ) would most likely be involved in the context of a partnership with SpaceX (SPCX) 4 hours ago

Ukraine President Zelensky says the Tesla (TSLA)/SpaceX (SPCX) Starlink and Starshield service remains slowed 4 hours ago

Meta (META) says it will extend the implementation date of updated commercial terms to November 20th (prev. October 30th); extension will allow it to explain the changes to partners and stakeholders 6 hours ago

Tesla (TSLA) has decided not to proceed with plans to produce Cybertruck for Australia and New Zealand 19 hours ago

[MARKET ANALYSIS] Crude benchmarks edge higher as the geopolitical environment remains tense 10 hours ago

Bristol Myers Squibb (BMY) spokesperson says the company has identified the cause of an earlier Car-T safety signal; will look to resume the paused trials in 2027 3 hours ago

Abbott (ABT) presents new ESMO data showing 30% absolute improvement in sensitivity with the whole-genome sequencing Oncodetect test in colorectal cancer 6 hours ago

Daily US Equity Opening News: SPCX explores access to 80K US cell sites; OpenAI nears USD 30bln funding at $1.25T valuation 6 hours ago

US BROKER MOVES: HESM upgraded at Raymond James; NCLH downgraded at JPM 6 hours ago