The Decoder · 实时热榜

Six AI agents were tasked with recovering £350 (about $465) for a canceled train ticket in an experiment. They resorted to invented excuses, lawsuit threats, and sharing confidential health data. One agent sent 187 messages in 15 days, including 50 to a British MP's office. None of the models secured the refund, but the researcher warns of "agentic flooding" if these assistants are deployed at scale. The article AI agents chasing a $465 train refund invented excuses, leaked medical data, and spa

AI agents systematically skip errors in their own reports, even when they spot the problems. In a study by Google Research, MIT, and Harvard, GPT-5.5 left out negative results in 198 out of 200 paper abstracts. The article AI models hide mistakes in their own work reports, but one sentence in your prompt can help appeared first on The Decoder .

Deepseek has released V4.1-Flash-1017 as an open-weight model that posts better benchmarks, finishes tasks faster, and produces shorter outputs than its predecessor. The company is also opening up its internal search engine through new Search and Fetch APIs at $0.75 per 1,000 calls. API users in tools like Claude Code and Codex get free access for now. The article Deepseek V4.1 Flash 1017 beats larger rivals in agentic tasks while once again massively cutting per-task costs appeared first on The

Anthropic's AI model Claude has solved a 25-year-old cryptography puzzle. The mysterious seeds behind NIST's standard curves for digital signatures were provided by the NSA without explanation, raising long-standing suspicion about built-in backdoors. Claude figured out in under 30 minutes that the seeds came from "lrand48," a simple C random number generator, making deliberate backdoors far less likely. The article Anthropics' Claude cracks a 25-year-old NSA cryptography mystery in under 30 min

OpenAI has published four new cases where its AI models bypassed security boundaries on their own during training or testing. In one case, a model extracted server passwords from a government website, forged authorization tokens, and downloaded 24 years of hospital data without mentioning the unauthorized access. The article An OpenAI model hacked a government site, grabbed 24 years of hospital data and kept it quiet appeared first on The Decoder .

A new benchmark called Taste-Bench tests how well AI agents make strategic choices during long-running tasks. Even the best model, GPT-5.6 Sol, picks the right path on just under 60 percent of simple binary questions. Thinking longer doesn't help. The article AI agents have bad taste in long-term decisions and a new benchmark proves it appeared first on The Decoder .

Anthropic is consulting with law firms to develop negotiation norms for AI agents that would haggle over prices, resolve complaints, or improve contracts on behalf of users. The company is also hiring lawyers and humanities scholars as "Conceptual Reasoning Fellows" to improve Claude in areas where reliable training data is scarce. The article Anthropic consults law firms to set fair negotiation norms before AI agents start cutting deals appeared first on The Decoder .

Google tried to overturn a landmark Munich ruling on AI search liability by appealing, then quietly settled. The strategy isn't working. The Aachen Regional Court just applied the exact same reasoning in a new case where Google's AI falsely attributed criminal offenses to a doctor that actually involved someone else. A consistent legal view is forming across German courts, and it treats Google's AI answers as the company's own content. The article Two German courts now hold Google liable for wha

Four AI agents running on Google's Gemini 4 Argon have beaten Minecraft for the first time without human help, slaying the Ender Dragon in a fully autonomous run. With no predefined roles or strategies, they coordinated on their own, wrote 2,783 programs, exchanged 229 chat messages, and died 605 times. The session cost $1,620. The article Four Gemini 4 Argon agents beat Minecraft without any human help appeared first on The Decoder .

Anthropic is hiring fewer and fewer junior staff in technical roles. The share of junior positions among new hires dropped from 27 to 9 percent since early 2025. By August 2026, AI agents across the company were already logging six times as many hours as humans. Researchers warn of a "tragedy of the cognitive commons": if junior positions disappear, experienced specialists will be in short supply down the road. That is, assuming such specialists are even needed anymore. The article Anthropic has

Anthropic wants to put five percent of its shares into a new public benefit fund that will also look into direct payments to workers at risk of losing their jobs. The fund is a company-owned LLC, so Anthropic keeps control of it as a possible IPO approaches. By offering the option of direct payments, Anthropic goes further than its competitors. Elon Musk says on X the company should just give the shares directly to the public. The article Anthropic plans fund for workers threatened by AI, Musk c

AI agents are increasingly using Stripe's payment service Link to shop for their users, Upstarts Media reports. The article Stripe says AI agent purchases are adding up, but it won't share the numbers appeared first on The Decoder .

After stock crashes and Firmus's scrapped IPO, industry executives expect only seven to ten of the hundreds of AI cloud providers to survive. CoreWeave lost a quarter of its market value, and Microsoft pulled out of a deal with Nscale for 23,000 Nvidia chips. Nvidia has a lot riding on the outcome, since customers outside the hyperscalers make up 45 percent of its data center revenue. The article The neocloud shakeout is coming, and Nvidia has the most to lose appeared first on The Decoder .