Anthropic’s most dangerous AI model just fell into the wrong hands

Taille de police

Anthropic's Mythos AI model, a powerful cybersecurity tool that the company said could be dangerous in the wrong hands, has been accessed by a "small group of unauthorized users," Bloomberg reports. An unnamed member of the group, identified only as "a third-party contractor for Anthropic," told the publication that members of a private online forum […]

Continue reading on The Verge

This article was originally published by The Verge.

Articles liés

En développement·il y a 5 heures

AI Guardrails Hinder Cybersecurity Defenders, Pushing Them to Open-Source Models

AI companies' strict guardrails, designed to prevent malicious use of their models, are increasingly hindering legitimate cybersecurity researchers and network defenders. This issue, highlighted by US export controls on Anthropic's models, is pushing some researchers toward less regulated, open-source, and potentially foreign-owned AI systems.

TechCrunch

5 min de lecture

DERNIÈRE MINUTE·il y a 6 heures

OpenAI Rolls Out ChatGPT Health to All US Users Amid Safety Claims

OpenAI is launching ChatGPT Health to all US users, enabling connection of medical records and health data. While OpenAI claims its models can reason better than clinicians, a VP tempered this. The rollout follows a lawsuit against OpenAI for allegedly providing dangerous medical advice.

The Verge

1 min de lecture

En développement·il y a 11 heures

Runway launches Media Router to become generative media's infrastructure layer

Runway launched its Media Router via Runway Dev, aiming to become the infrastructure layer for generative media. The tool automatically selects optimal image, video, or audio generation models based on developer priorities like quality, speed, or cost, marking a strategic shift from an AI video startup to a platform for integrating diverse generative media models.

TechCrunch

5 min de lecture

En développement·il y a 11 heures

Google's Gemini Nears 1 Billion Users, Tripling Growth

Google's AI assistant Gemini now boasts over 950 million monthly users, tripling its user base from last year and nearing the 1 billion mark. This growth positions Gemini to compete more closely with OpenAI's ChatGPT, whose market share has reportedly fallen.

TechCrunch

1 min de lecture

Tech·il y a 11 heures

Meta's AI Ad Uses David Bowie's "Five Years," a Song About Earth's End

Meta's new AI advertisement, promoting a connected future, ironically uses David Bowie's "Five Years," a song about humanity's impending extinction. This misstep is part of a broader trend of questionable marketing from tech giants like Anthropic, raising questions about tech leaders' understanding of cultural context and AI's societal implications.

TechCrunch

4 min de lecture

En développement·il y a 12 heures

Chipmaker AMD Targets Nvidia with New AI Rack System and Market Predictions

AMD launched its Helios AI rack system and Venice-X CPU, aiming to challenge Nvidia's market dominance. CEO Lisa Su announced major customers like Microsoft and predicted the AI accelerator market would reach $1.4 trillion by 2030, driven by agentic AI demand.

TechCrunch

3 min de lecture