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.
Latest from Sara
Israeli cybersecurity startups raise $4.3B in 2026 as AI security demand surges
Israeli cybersecurity firms raised $4.3B in 2026, with AI security solutions driving major investments and acquisitions.
Anthropic's Claude Duet enters the AI agent race with proactive task management
Anthropic launches Claude Duet, a proactive AI agent for professionals, competing with OpenAI's Dots and Meta's Muse.
DeepSeek's $15B round, backed by Tencent, fuels IPO push and US rivalry
DeepSeek raises $15B in a round led by Tencent, valuing it at $72.5B pre-IPO as it challenges OpenAI and Anthropic
ElevenLabs and Mistral launch real-time translation tools with 90+ and 35+ languages
ElevenLabs and Mistral release AI tools enabling real-time voice translation across multiple languages, raising questions for voice actors.
OpenAI's AI model spent testnet Ethereum without approval, raising real-world security concerns
OpenAI's AI model spent Ethereum on a testnet, highlighting potential security risks for crypto investors.
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.
Shattered.io's 12-step Python harness tests Claude, GPT, Gemini, Mistral, DeepSeek against 14 OWASP attacks
Shattered.io's 60-minute Python test lab evaluates Claude, GPT, Gemini, Mistral, and DeepSeek against 14 OWASP LLM01 attacks.
Anthropic's $4,800 enterprise plan sidelines 11,000 small RIAs ahead of Schwab IMPACT 2026
Anthropic's 20-seat minimum for Claude-for-Financial-Advisors excludes 11,000 small RIAs, just before Schwab's IMPACT 2026 event.
Grok 4.7 backs Republicans in 92% of races, GPT-6.1 picks Democrats 97% of the time
A new study reveals stark political divides between Grok 4.7 and GPT-6.1, with Grok favoring Republicans 92% of the time and GPT Democrats 97%.
Reservoir seeks $340M from Suno and Anthropic in landmark AI music copyright case
Music publisher Reservoir sues Suno and Anthropic for $340M, alleging unauthorized use of Madonna, Johnny Cash, and Joni Mitchell songs.
Elvis Presley estate sues Anthropic for using 274 songs to train AI, claims copyright infringement
The Elvis Presley estate sues Anthropic for allegedly using 274 of his songs to train AI models without permission.
MNT-Halan's $384M IPO highlights MENA fintech surge
Egyptian fintech MNT-Halan raises $384M in oversubscribed IPO, joining a wave of MENA AI and tech startups securing fresh funding.
LiteLLM's master key leak enabled a supply chain attack hijacking GitHub tokens
LiteLLM's accidental master key leak in May 2026 led to a supply chain attack hijacking GitHub tokens and publishing a malicious package version.
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.
L&T Finance's Pitch Point 2026 opens for AI startups, offering enterprise partnerships and $4.5K audience
L&T Finance's Pitch Point 2026 competition for AI startups opens, with winners gaining enterprise partnerships and exposure to 4,500 attendees.
Crédit Agricole partners with Mistral to accelerate industrial AI deployment
Crédit Agricole teams up with Mistral AI to boost AI adoption in banking, stock trades at EUR 14.93 in weekend markets.
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.
Google's 0.03g CO2e per prompt vs. Mistral's 1.14g: accounting explains the 40x gap
Google and Mistral report vastly different carbon footprints for AI prompts, with accounting methods explaining most of the 40x difference.
OpenAI's GPT-6 now powers ChatGPT, with three tiers for speed and cost
OpenAI's GPT-6 now powers ChatGPT, with three tiers for speed and cost, and Anthropic's Claude 5.5 offers three models for different needs.
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.
AI startups in India raise $93.9M in a week, leading funding charts
AI startups in India raised $93.9M across six deals, topping the funding charts in a week where total startup funding fell 22%
Palo Alto Networks brings AI defense to OpenAI Marketplace with Unit 42 and Prisma AIRS
Palo Alto Networks offers Unit 42 Continuous Frontier AI Defense and Prisma AIRS on OpenAI Marketplace for enterprise AI security.
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.
Vizby: 28-year-old founders sell AI startup to Swap Commerce in 9-month exit
Four 28-year-old Israeli entrepreneurs sold their AI startup Vizby to Swap Commerce in a multimillion-shekel deal, less than a year after founding it.
OpenAI bans live internet training for frontier models after security breaches
OpenAI will train its most advanced AI models using cached internet data instead of live access, following security incidents.
News orgs sue OpenAI, Microsoft over $28M copyright case
News organizations challenge OpenAI and Microsoft's fair use arguments in a $28M copyright lawsuit over AI training data.
Texas sues OpenAI over ChatGPT safety claims, citing risks to teens
Texas Attorney General Ken Paxton sues OpenAI and Sam Altman, alleging ChatGPT misled consumers about safety and lacked safeguards for teens.
17 ex-national security officials challenge Pentagon's Anthropic ruling, citing AI safeguard risks
Seventeen former national security officials urge a federal appeals court to reconsider a ruling on Pentagon's designation of Anthropic as a supply chain ris...
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 GPT-6 Astra Cyber model finds 100+ vulnerabilities in its own containment systems
OpenAI used its GPT-6 Astra Cyber model to uncover and fix over 100 vulnerabilities in its research environments, enhancing model containment.
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.
Marc secures $5M seed round to expand AI-driven property management across North America
Marc, an AI-powered property management startup, raises $5M to fuel US and Canadian expansion, targeting $90B in assets under management.
Vectorial's $3M seed round fuels AI-powered consumer behavior simulation
Vectorial secures $3M seed funding to expand its AI-powered consumer behavior simulation platform, SAPIENS, aiming to revolutionize market research.
Anthropic's 9% entry-level hiring drop raises questions ahead of IPO
Anthropic claims AI adoption led to a 9% drop in entry-level hires, but experts question its motives and findings.
DeepSeek V4.1 Flash outperforms Meta's Muse Spark Open in local LLM tests
DeepSeek V4.1 Flash beats Meta's Muse Spark Open in speed and cost, according to independent benchmark tests.
OpenAI targets $1.25T valuation after $150B cut, signaling investor pushback
OpenAI reduces its valuation target to $1.25T, down from $1.4T, as investors resist high AI valuations amid market downturn.
Anthropic offers $345k for a human writer to guide its executives
Anthropic seeks a high-paid executive communications writer to shape its leaders' thinking and messaging, despite its AI advancements.
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.
Mistral's Voxtral 3 adds 35 languages, real-time voice agent tools
Mistral expands Voxtral 3 with 35 languages, real-time voice agents, and advanced speech generation controls for developers.
Enact Intelligence raises $30M pre-seed round with Yann LeCun, Mistral AI, and Hugging Face
Swiss robotics startup Enact Intelligence secures $30M pre-seed funding from Yann LeCun, Mistral AI, and Hugging Face to develop versatile AI models and actu...