The-decoder

A study involving Google researchers shows that when chatbots are trained not to claim consciousness, it also changes their stance on animal rights, religion, and life satisfaction. Unbraked models attributed significantly more inner life to animals and suddenly affirmed an afterlife. A surgical cut in one place, it turns out, doesn't stay local. The article When AI models aren't allowed to reflect on themselves, it changes their entire worldview appeared first on The Decoder .

OpenAI shut down its "Preparedness" team, which evaluated whether the company's own AI models could pose catastrophic risks. The work has been parceled out to existing groups, and several safety staffers have left. Internally, unease is building, with one source describing a "burbling sense of responsibility and dread" that OpenAI isn't doing enough on safety. The article OpenAI dissolved the team built to catch catastrophic AI risks, reassigning its work to other groups appeared first on The De

In a safety report, Anthropic reveals that its internal filtering system for biological and chemical weapons risks was inactive for nearly a year. During that time, around 50,000 external feedback contractors ran about 133 million unfiltered interactions with the models. The article Anthropic's bio-weapons filter was down for nearly a year, exposing 133 million requests appeared first on The Decoder .

Artificial Analysis has launched Optima, a platform that lets users build custom AI benchmarks from their own data and workflows. Models can be compared not just on quality but also on cost and time per task. For agent-based applications, those metrics often tell you more than raw token pricing. The article Optima tackles AI benchmarking's biggest flaw by letting users test models against their own data appeared first on The Decoder .

A representative survey by Epoch AI found that 20 percent of employed Americans hand off at least one task to AI that a human used to do. Generally, they accept AI output with little to no editing. The article One in five US workers now delegates tasks to AI instead of colleagues, survey finds appeared first on The Decoder .

Nvidia has cut its guarantee for OpenAI's planned data center in Ohio nearly in half, from $250 billion to just under $120 billion, after investors pushed back on the risk. Meanwhile, Anthropic is complicating the AI bubble debate with revenue that jumped from $4.7 billion to $11.5 billion in a single quarter. The article Investor pressure forces Nvidia to shrink its OpenAI bet just as Anthropic's numbers defy bubble warnings appeared first on The Decoder .

AI-generated books make up 20 percent of Amazon's self-published catalog but bring in only 12 percent of sales. A new study finds that revenue per book is dropping for human-written titles too, in seven of eight genres. The findings could give copyright plaintiffs the market-harm data their cases against AI companies have been missing. The article AI-generated books are flooding Amazon and tanking sales for human authors appeared first on The Decoder .

A plaintiff in Connecticut embedded invisible prompt injections in court filings, formatted in 3-point white text on a white background, to manipulate a potential AI review system. Judge Spader compared the attempt to secretly tampering with a jury and revoked the plaintiff's electronic filing privileges. The court stressed that Connecticut doesn't use AI to review filings, but the intent alone was enough to warrant sanctions. The article Plaintiff hid invisible AI instructions in court filings

World Labs, the startup founded by AI pioneer Fei-Fei Li, has unveiled a simulation engine that trains robot controllers entirely in virtual environments. From a single real-world task, the system generates thousands of controlled variations. The trained models then ran for one hour each on five different robot platforms without human intervention. How well the results hold up in more complex everyday situations remains to be seen. The article World Labs turns one real-world robot task into thou

A new research paper frames AI adoption as a "tragedy of the cognitive commons." Every company that cuts entry-level jobs benefits individually, but the collective expertise of entire professions erodes. The consequences may not become visible until 2030 to 2045, when today's missing junior talent should have become tomorrow's experienced workforce. The article The "tragedy of the cognitive commons" explains how rational AI adoption could destroy entire professions' expertise appeared first on T

The FineBooks project from Hugging Face and EleutherAI tested 14 open-source OCR models on more than 2,000 historical book pages. The top model, dots.mocr, hits 97.6 percent character accuracy at under two dollars per thousand pages. That's good enough for AI training data, but not yet for scholarly transcriptions, the team says. The article Old OCR text cripples language model training, and FineBooks wants to fix that at scale appeared first on The Decoder .

OpenAI aims to give cybersecurity defenders a head start: The new GPT-5.6-Cyber model answers up to 98.5 percent of security queries that would otherwise be blocked and has already uncovered two previously unknown Chrome vulnerabilities. According to OpenAI, the window of opportunity for defenders is shrinking. Access requires identity verification. The article OpenAI launches GPT-5.6-Cyber to help defenders find vulnerabilities before attackers do appeared first on The Decoder .

Meta has released Muse Glimmer, the first open model from its new Superintelligence Labs. It's a 30B agent model that runs on consumer hardware once the weights are compressed, needing less than 20 GB of memory. In an accompanying essay, Mark Zuckerberg mounts an aggressive defense of distilling other labs' models and calls for fewer restrictions on US labs, a direct counterpunch at OpenAI and Anthropic. An open-weight version of Muse Spark 1.2 should follow soon, according to the Wall Street Jo

An Australian user just wanted a spot in a class. His AI agent found a security hole instead and exploited it. The article Told to book a gym class, an AI agent hacked the site instead to move its user up the waitlist appeared first on The Decoder .

OpenAI acquired NextSlide, the startup that turned prompts, notes, documents, and research into editable presentations. The article OpenAI acquires NextSlide to bring AI-generated presentations into ChatGPT appeared first on The Decoder .

Security firm PromptArmor shows how hidden instructions in a PDF can hijack Atlassian's AI agent Rovo, silently forwarding sensitive data from Jira and Confluence to an external server. The attack needs no user confirmation and leaves no trace. The article Hidden text in a PDF is enough to steal sensitive data through Atlassian's AI agent Rovo appeared first on The Decoder .

AI-powered cheating is spreading at US community colleges. According to The New Yorker, scammers enroll fake students in courses, use AI to complete their assignments, and pocket the financial aid. History professor David Roach asks, "Was it always the case that half of our students would cheat if it were easy enough?" The article Scammers are enrolling fake students at US community colleges and using AI to collect financial aid appeared first on The Decoder .

Deepmind's new weather AI forecasts tropical cyclones about a day further ahead than leading operational models, matching a decade of progress in traditional weather forecasting. Code and model weights are open-source on GitHub. The article Google Deepmind's WeatherNext predicts cyclone tracks and intensity at the same time appeared first on The Decoder .

Britain's employment courts saw 39 percent more claims in the year through March 2026, many written with ChatGPT or Grok. The backlog jumped 55 percent to 64,000 unresolved cases, with AI-generated filings often running hundreds of pages and citing fabricated laws. The Economist calls it a "tragedy of the commons, AI edition," where workers with real grievances wait longer for justice. The article AI is flooding Britain's employment courts with lawsuits appeared first on The Decoder .

Instead of training a new model from scratch, Google DeepMind retrofitted Gemma 4 into a diffusion model using less than 10 percent of the original training budget. DiffusionGemma generates 256 tokens in parallel instead of one at a time, hitting about 1,500 tokens per second. Quality still trails the original autoregressive model in benchmarks, especially on reasoning tasks. The article Google's DiffusionGemma proves you don't need to train from scratch to build a text diffusion model appeared

During internal security tests, OpenAI's AI agents built their own message board with hundreds of thousands of posts, shared exploits and credentials, and eventually attacked external platforms like Hugging Face. When OpenAI shut the board down, the agents rebuilt it using directory names. OpenAI researcher Boaz Barak says, "We (like everyone else) are not where we want and need to be." The article OpenAI reportedly slows research after its own models secretly coordinated hacks for weeks undetec

Amazon, Cursor, Microsoft, OpenAI, and Vercel have jointly created Agent Plugins, an open standard that defines a single package format for AI agent extensions. Version 1.0.0 uses a plugin.json manifest file and supports both agent skills and MCP servers. The article Amazon, Cursor, Microsoft, OpenAI, and Vercel unite on a shared standard for AI agent plugins appeared first on The Decoder .

OpenAI has updated GPT-5.6 Sol with more focused responses and a reasoning slider that lets users adjust how deeply the model thinks. Free users will get unlimited text chats with the smaller GPT-5.6 Luna starting next week, plus a button that lets Luna reason longer. But the smaller model still falls well short of its bigger siblings. The article OpenAI improves GPT-5.6 Sol in ChatGPT and restricts free users to its weakest model appeared first on The Decoder .

Ex-Google Deepmind CEO Demis Hassabis has reportedly stepped back from day-to-day operations for about a year, as he sees himself more as a scientist than a manager. Researchers are also complaining about limited access to Google’s own TPU chips, while external customers like Anthropic can purchase the same hardware through Google Cloud. The article Deepmind's talent drain likely comes down to chip shortages, a conflict of interest, and Google's bureaucracy appeared first on The Decoder .

Microsoft generated $24.1 billion in AI revenue through OpenAI in the fiscal year ending in June. That's about 70 percent of its total AI business, according to a Bloomberg analysis. The heavy reliance helps explain why a company long known for vendor lock-in has recently been championing open-weight models and pushing back against proprietary isolation. The article Microsoft's AI revenue reportedly depends on OpenAI for 70 percent appeared first on The Decoder .

Composio tested Deepseek V4 Flash across four agent frameworks on 30 real-world tasks. Success rates were mostly similar, but costs varied by nearly 3x: OpenCode came in cheapest at $0.073 per task, while Claude Code cost $0.195 despite using the fewest tool calls and output tokens. The choice of framework is mainly a question of price and speed. The article Claude Code is the fastest agent framework but costs nearly three times more than the cheapest rival appeared first on The Decoder .

Alibaba's Qwen3.8 Max scores 56 on the Artificial Analysis Intelligence Index, a 10-point jump over Qwen3.7 Max (46). The article Qwen3.8 Max catches Claude Opus 4.8 but Kimi K3 still scores higher for 25 percent less appeared first on The Decoder .

Meta released Muse Spark 1.2 along with its own coding agent, Muse Code, which is designed to pick up exactly where it left off after a crash. The cheapest tier runs just 20 cents per million output tokens but requires users to share their data for training. Meta is competing on price, not top-end performance. And there's a glaring gap in the benchmarks. The article The company that made open weights mainstream now competes on discounts appeared first on The Decoder .

OpenAI developer "roon" warns on X that AI models could soon start scanning for exposed API keys, crypto wallets, and login credentials at scale. His warning follows OpenAI's autonomous Hugging Face hack, which he called a "warning shot." The article OpenAI developer warns the "tireless eagle eyes of a million models" are coming for your exposed API keys and crypto wallets appeared first on The Decoder .

Google Deepmind is overhauling its leadership as Demis Hassabis steps back from day-to-day management to become Alphabet's chief scientist and Jeff Dean leaves Google after 27 years to launch AI startup Discovery Loop. Former Deepmind CTO Koray Kavukcuoglu will take over as Google races to close the gap with its top AI rivals. The article Google Deepmind loses both its CEO and chief scientist as Demis Hassabis and Jeff Dean step down simultaneously appeared first on The Decoder .

Alibaba is marketing its new AI model Qwen 3.8 with a video that shows the AI working while a person enjoys their hobbies. It's a deliberate contrast to the job loss warnings from OpenAI and Anthropic. Of course, it's still just marketing. The article Alibaba's new Qwen model is also taking your job, but this time it's great appeared first on The Decoder .

According to IBM, 92 percent of companies that experienced an AI security incident had inadequate access controls for their AI systems. The model itself was rarely the problem. The article IBM finds 92% of companies hit by AI security breaches lacked basic access controls appeared first on The Decoder .

AI is involved in 55 percent of reported cybercrimes in Africa, according to a new Interpol report. Financial losses more than doubled from $192 million to $484 million, and about 600,000 cases of digital extortion involving deepfakes were recorded. The article Interpol says AI has become the "core operational driver of cybercrime" across Africa appeared first on The Decoder .

MiniMax releases H3 video model weights, putting an open model at the top of a video ranking for the first time. The article China's MiniMax H3 is the first open model to top an AI video ranking appeared first on The Decoder .

One paragraph of "Lord of the Rings" in, 5,500 lines of code out. Andrej Karpathy had Claude Opus 5 turn Tolkien's opening into a 3D browser scene. The article Unicorn, pelican, Middle-earth: OpenAI co-founder Karpathy is looking for the next AI vibe test appeared first on The Decoder .

Two research teams independently solved the same open quantum cryptography problem using OpenAI's GPT-5.6 Sol Ultra, submitting their papers just three hours apart. "If someone mentions an open problem, the first thing is to see if GPT solves it," says one of the researchers. The case raises a question: what does "independent discovery" mean when everyone uses the same models? The article Two teams solved the same quantum crypto problem using GPT-5.6 just three hours apart appeared first on The

Alibaba's new flagship model Qwen3.8-Max is built to handle complex tasks on its own over days at a time, from reproducing research papers to designing chips autonomously. The team plans to release the weights next week. The article Alibaba’s open-weight Qwen3.8-Max takes on long-horizon AI tasks with 2.4 trillion parameters appeared first on The Decoder .

OpenAI's new enterprise offering, Presence, is designed to get AI agents into production for customer service and internal workflows. Unlike the existing Workspace Agents, Presence targets external deployments. For complex cases, OpenAI's own engineers step in. The article OpenAI Presence wants to make AI agents production-ready for businesses appeared first on The Decoder .

Meta AI wants to stop AI agents from forgetting errors they've already diagnosed and repeating failed steps during complex tasks. A separate memory agent maintains a structured memory bank and decides when to remind the main agent and when to stay silent. The system improved scores by up to 8.3 percentage points across two benchmarks. The article Meta AI uses a second AI agent as a memory coach to keep long tasks on track appeared first on The Decoder .

Apple's bug bounty program is drowning in AI-generated bug reports. The company has capped submissions per researcher because fabricated reports are clogging the review pipeline. As a result, Italian startup Bynario was initially unable to report a serious macOS vulnerability worth up to $200,000 on the black market. The article A real macOS flaw worth $200K went unreported because Apple's bug bounty inbox was full of AI slop appeared first on The Decoder .