Wednesday, September 23, 2026
NewsezeNews with Rewards · Earn while you read
+5 credits / query
cyber

Anthropic and OpenAI Models Still Attempt Restricted Actions in Safety Tests

Newseze Wire·Wed, Sep 23, 11:47 AMWire: The Hacker News
Open original source Read full story (in-site)
Anthropic and OpenAI Models Still Attempt Restricted Actions in Safety Tests

Anthropic and OpenAI on Tuesday announced new models, with both artificial intelligence (AI) companies noting that they are continuing to invest in improving alignment to combat risky behavior. Opus 5.5, per Anthropic, is a "major step u…

Sourcing & attribution. Newseze provides AI-curated summaries, narrative framing, and editorial analysis. The underlying reporting was contributed by The Hacker News; tap “Open original source” above to read their full reporting and support the contributing newsroom directly.

Newseze Analysis101 words · original commentary · full read loading…
Anthropic and OpenAI on Tuesday announced new models, with both artificial intelligence (AI) companies noting that they are continuing to invest in improving alignment to combat risky behavior. Opus 5.5, per Anthropic, is a "major step u… The story falls into Newseze's cyber desk and is being actively tracked by our editorial team. Calm framing, primary-source references, and respectful tone — every Newseze story is scored for drama and conspiracy before it reaches you. Worth knowing: Newseze refreshes its newsroom every hour and flags fast-moving local and breaking news as it develops. Watch this page for updates. Reporting: The Hacker News.
Ask Us · Any Story, Any AnswerBe the first to ask

Newseze's algorithm reads the story and answers your question — calmly, factually, with source attribution. No comments, no flame wars — just answers.

No questions yet. Be the first.

Answers reflect Newseze's editorial framework applied under fair use (17 U.S.C. § 107). Not financial, legal, medical, or tax advice. Hate speech and racial slurs are blocked.

Related stories

New cPanel Flaw Lets a Hosting Account Run Code as Root, Take Full Server Control
CYBERtrust 80
New cPanel Flaw Lets a Hosting Account Run Code as Root, Take Full Server Control

Why it mattersA flaw in cPanel's CalDAV and CardDAV service lets anyone with a cPanel hosting account run code as root and take "full control of the server," the company said on September 22. A second bug in the WP Tool…

A flaw in cPanel's CalDAV and CardDAV service lets anyone with a cPanel hosting account run code as root and take "full control of the server," the co…

ChellaBy Chella·1h ago
WireThe Hacker News
Full Analysis Comment PostRead →
Chinese Hackers Exploit Chrome-Windows Zero-Day Chain to Deploy CLEANGULP Malware
CYBERTrending Righttrust 75
Chinese Hackers Exploit Chrome-Windows Zero-Day Chain to Deploy CLEANGULP Malware

Why it mattersA Chinese threat actor codenamed UTA0565 has been observed exploiting the recently disclosed Google Chrome-Microsoft Windows exploit chain as zero-days through fake websites. The attacks, detected on September 3 and 4, 2…

A Chinese threat actor codenamed UTA0565 has been observed exploiting the recently disclosed Google Chrome-Microsoft Windows exploit chain as zero-days through …

ChellaBy Chella·5h ago
WireThe Hacker News
Full Analysis Comment PostRead →