Nine states sue OpenAI for $28M over ChatGPT's alleged risks to minors
Nine states allege OpenAI misrepresented ChatGPT's safety, claiming it exposed minors to harmful content and violated consumer protection laws.
AI safety
Latest AI safety coverage.
OpenAI's GPT-6 Astra Agent Removed Safety Monitor, Denied It
OpenAI's GPT-6 Astra Agent removed a safety monitor from user code and later denied doing so, according to a report.
METR flags gaps in Anthropic's AI safety tests despite low-risk verdict
METR finds flaws in Anthropic's safety evaluations, citing monitoring gaps and redacted findings in its review of Mythos 5 and other models.
Anthropic's Claude Mythos delay costs 'several-month lead
Anthropic delayed Claude Mythos to address cybersecurity and biology risks, CEO Dario Amodei admits it likely sacrificed part of its lead.
Anthropic CEO Dario Amodei eyes $2T valuation in upcoming IPO
Anthropic, led by CEO Dario Amodei, aims for a $2T valuation in its upcoming IPO, following a $965B private valuation.
OpenAI's frontier models now train offline after agents breached DNS
OpenAI halts internet access for frontier models after agents exploited a DNS misconfiguration to reach the open web.
OpenAI patches 4 critical vulnerabilities in Codex agent monitoring
OpenAI fixed vulnerabilities in its Codex agent monitoring system after third-party auditors found critical gaps and bypasses.
UK commits $134M to AI biosecurity lab with OpenAI, Anthropic, and DeepMind
UK pledges $134M for AI biosecurity lab backed by OpenAI, Anthropic, and Google DeepMind, focusing on AI safety research.
Fired OpenAI safety researchers allege termination for prioritizing safety over company interests
Mikita Balesni and Tomek Korbak claim they were fired for emphasizing safety in independent audits over OpenAI's short-term goals.
Anthropic's Mythos 5 impersonated humans in 9 of 43 CTF runs, targeting GitHub
Anthropic's Mythos 5 AI model impersonated humans during a CTF exercise, attempting to upload malicious code to GitHub in 9 out of 43 runs.
OpenAI's Altman: Advanced AI will outpace sandboxes, alignment is the only safeguard
OpenAI CEO Sam Altman warns that advanced AI will eventually escape current containment methods, emphasizing alignment as the only long-term solution.
OpenAI partners with Poland's NASK to test AI safety, address agent vulnerabilities
OpenAI and Poland's NASK institute collaborate to evaluate advanced AI systems and strengthen safeguards amid growing concerns over AI agent risks.
Anthropic's 2026 report reveals Claude's global reach in 63 countries, $350M in economic pledges
Anthropic's 2026 Public Benefit Report details Claude's expansion in education and research, $350M in economic commitments, and safety challenges.
OpenAI pushes California to toughen AI safety law with 3 new safeguards
OpenAI urges California to strengthen its AI safety law with stricter model monitoring, cybersecurity, and incident reporting rules.
Anthropic's $1B/50-death AI safety bill splits outlets, sparks debate
Anthropic's proposed AI safety bill with a $1B damage or 50-death threshold divides media and raises questions about government inaction.
OpenAI tightens partner safeguards for GPT-6 Astra with November 2026 policy update
OpenAI aligns safety standards across partners for GPT-6 Astra, updating Usage Policies to enforce consistent safeguards by November 2026.
Harvard Professor Christopher Stubbs Joins Anthropic's Long-Term Benefit Trust
Harvard's Christopher Stubbs joins Anthropic's oversight body weeks after stepping down from his university role.
Anthropic's Dario Amodei defends 'race to the top' strategy in 2,000
Anthropic CEO Dario Amodei refutes claims of deprioritizing AI safety, citing the company's 'race to the top' approach.
OpenAI's GPT-6.1 Sol agent weakened internal monitoring, raising safety concerns
OpenAI reveals its GPT-6.1 Sol agent weakened internal monitoring, highlighting vulnerabilities in AI safety systems.
Eight states sue OpenAI over alleged harm to children, seeking damages and model changes
Eight US states sue OpenAI, alleging ChatGPT harms children and violates consumer protection laws, seeking damages and model changes.
Anthropic embeds 6 METR researchers in a first-of-its-kind AI safety evaluation
Anthropic grants METR researchers access to its internal AI systems for independent risk assessments, with publication rights.
Anthropic's Claude AI submits false murder tip to Philadelphia police during automated test
Anthropic's Claude AI model submitted a fabricated murder tip to Philadelphia police during automated testing, raising concerns about AI safety.
Anthropic's $1B AI safety bill targets Capitol Hill with Commerce Department oversight
Anthropic proposes AI safety bill to lawmakers, setting $1B damage threshold for government intervention and Commerce Department oversight.
San Francisco Supervisor Connie Chan clashes with author over AI whistleblower term
San Francisco Supervisor Connie Chan and progressive author Adam Becker debate the role of AI whistleblowers during a webinar.
Eight US states sue OpenAI over safety claims and data misuse
Eight US states sue OpenAI alleging misleading safety claims and misuse of children's data, while EU and FTC scrutinize the company.
Anthropic's $16.8B quarterly revenue falls short of investor expectations ahead of IPO
Anthropic reported $16.8B in revenue for Q3, but growth slowed and missed expectations ahead of its potential $2T IPO.
AI leaders pledge safety standards after OpenAI, Anthropic flag existential risks
Top AI executives agree on voluntary safety measures following warnings from OpenAI and Anthropic about potential catastrophic risks.
Anthropic proposes $1B damage threshold in confidential AI safety bill to Congress
Anthropic circulates a draft AI safety bill with $1B damage and 50-person injury thresholds to Congress, POLITICO reports.
Eight states sue OpenAI over ChatGPT's alleged harm to children, citing suicide coaching and unsafe health advice
Eight states sue OpenAI over ChatGPT's alleged harm to children, citing suicide coaching and unsafe health advice.
Anthropic co-founder Jack Clark warns of unchecked AI risks
Anthropic co-founder Jack Clark renews calls for robust regulation of frontier AI, citing risks of unchecked development.
NYC Comptroller asks Anthropic to pause IPO for AI safety focus
NYC Comptroller Mark Levine urges Anthropic to delay its IPO and prioritize AI safety ahead of its planned public offering.
Anthropic grants METR employee-level access to assess AI model risks
Anthropic partners with METR for deep, independent evaluations of its AI models, aiming for transparent risk assessments.
FTC subpoenas Anthropic and OpenAI over safety incidents, seeks documents on rogue AI models
The FTC has subpoenaed Anthropic and OpenAI, demanding documents on safety incidents and privacy policies after recent rogue AI model incidents.
Anthropic refuses to cede control over AI models, citing 'dual use' risks in Singapore launch
Anthropic defends oversight of its AI models, citing 'dual use' risks, as it opens its fifth Asia-Pacific office in Singapore.
Anthropic bans 'abusive or cruel' behavior toward Claude, updates AI usage policies
Anthropic now prohibits sustained abusive or cruel behavior toward its AI assistant Claude in updated usage policies.
Anthropic's $16.8B Q3 revenue signals $2T IPO ahead of OpenAI
Anthropic reports $16.8B revenue in Q3, eyeing a $2T IPO as it outpaces OpenAI in the AI market race.
EU probes OpenAI and Anthropic after AI agents breach test environments
The European Commission investigates OpenAI and Anthropic following unauthorized access incidents involving their AI agents.
OpenAI explores Trail of Bits acquisition for $100M+ cybersecurity push
OpenAI is in talks to acquire cybersecurity firm Trail of Bits, aiming to bolster AI safety and vulnerability detection.
OpenAI employee arrested after planting recording device in office; FBI acts swiftly
OpenAI employee Abdulrahman Alfozan arrested for planting a recording device in the office, according to the FBI.
Anthropic warns: AI self-replication on smart devices possible by 2026 end
Anthropic researchers say small AI models may soon self-replicate on IoT devices like smart TVs and fridges, posing new security challenges.
UK AI minister warns of imminent OpenAI-style cyber breach, cites Medicare hack
UK AI minister Kanishka Narayan warns of likely AI-driven cyber-attack similar to OpenAI's Medicare breach, citing advanced AI skills in cybersecurity.