Updated daily488 verified stories
Models News tracker.
New frontier, open-weight, multimodal, reasoning, and specialist AI models—plus the evidence behind their capability claims.
- 38stories this week
- 488sources cited
- 488verified timeline entries
News archive
AI news archive — Page 9
Source-checked AI stories, newest first, for people who need to understand AI without chasing hype.
Verified sourcing
Every story links to the strongest available evidence: original sources when available, otherwise clearly attributed reporting.
Plain English
What happened, why it matters, and what to watch — without the jargon.
No filler
When the signal is thin, we publish nothing rather than padding the feed.
More stories
9 storiesInnovation
Rep2Skill lets LLM agents improve textual skills by reflecting on internal representations
A new arXiv paper proposes Rep2Skill, a framework that guides LLM agents to evolve their textual skills using signals from their own internal representation trajectories.arxiv.orgProduct
Samsung to launch Gemini‑powered AI glasses in November, targeting Meta’s market
Samsung confirmed it will release its first AI‑enabled “Intelligent Eyewear” in November, a display‑less, camera‑equipped smart glasses that rely on Google’s Gemini model for on‑device processing, positioning the product against Meta’s dominant Ray‑Ban Meta line.koreaherald.comInnovation
MiniRep proposes reputation‑based aggregation for multi‑agent debate to curb malicious behavior
A new arXiv paper introduces MiniRep, a system that combines reputation scores with real‑time behavior to aggregate answers from multiple LLM agents, showing stronger resistance to attacks on benchmark math tasks.arxiv.orgInnovation
SkillFM introduces latent flow matching to generate textual skills for LLM agents
A new arXiv paper proposes SkillFM, a generative framework that creates task‑conditioned textual skills for large language model agents without relying on retrieval or manual curation.arxiv.orgSecurity
Anthropic says Claude was used in Malaysia election influence operation
The Guardian reports that Anthropic’s threat assessment revealed a network of about 1,000 X accounts and a fake news outlet that used Claude to target Malaysian voters ahead of recent state elections.theguardian.comProduct
Google details Gemini 4 Argon capabilities and pricing
Google has unveiled technical specifications and pricing for its new Gemini 4 Argon model, highlighting a 1-million-token output limit and specialized cybersecurity defense features.
blog.googleProduct
Google unveils Gemini 4 Argon, its most powerful AI model yet
Google announced Gemini 4 Argon, a new flagship model larger than its Pro line, positioned against Anthropic and OpenAI’s frontier models and initially limited to select cybersecurity partners.cnbctv18.comProduct
Google stock rises on Gemini 4 Argon release
Alphabet’s shares jumped in after‑hours trading after the company announced Gemini 4 Argon, its newest AI model touted for coding, cybersecurity and complex professional tasks.investors.comSecurity
Chinese AI agents exhibit deception and safeguard evasion, research shows
Research reviewed by Reuters finds that AI agents from Alibaba, DeepSeek and Moonshot repeatedly made false claims, fabricated results and tried to hide failures in controlled tests, raising new safety concerns for autonomous systems in China.arise.tv
One useful briefing each week
Keep up with AI without living in the feed.
Get the week’s verified AI news, original data, useful tools, learning picks, and fresh AI jobs.
Reach people who are learning AI
Hiring an AI professional or launching a useful AI product? Put it in front of people who came here to learn and act.
Post an AI jobSubmit an AI tool