best ai autonomous agent 2026

Copy link Facebook X Whatsapp Reddit Pinterest Flipboard Email Share this article Join the conversation Add us as a preferred source on Google Newsletter Subscribe to our newsletter Researchers have managed to execute code within an "llms.txt" file that many large companies use to instruct AI agents on how to scrape the website correctly. Back when the internet exploded and search engines became popular, sites started publishing a "robots.txt" file to guid

Meta / Futurism Meta’s platforms are in a sorry state as grotesque AI slop continues to proliferate widely. As a recent Futurism investigation found, many Facebook accounts have been flooding the platform with horrid videos depicting violent child abuse, and Meta appears to be woefully unprepared — and downright unwilling, judging by its ambiguously-phrased policies — to do anything meaningful about it. At the same time, the Mark Zuckerberg-led company is

Life, it seems, wants to play. Anyone who’s ever spent ten minutes with a five-year-old knows this perfectly well: very young children are masters at make-believe. And it’s not just people—play comes instinctively across the animal kingdom. Rats are fond of rough-and-tumble play, especially after periods of isolation. Octopuses will toy with Legos in much the same way that a newborn will play with blocks. Far from being a useless pastime, play appears to b

An agent completes a task in 30 steps: it calls the model, searches, writes to the database, and calls the model again. Suppose each step fails once in a hundred: the network flickered, the API returned “overloaded.” Then the task will run to completion in 74% of cases. One in four will fail. With a 3% failure rate per step, only 40% will run to completion.For this calculation, step failures are assumed to be independent and transient.In the demo, the agen

Published on September 18, 2026 at 12:45 Comment 0 𝕏 Print © 01net Entrusting your card to a stranger remains a bad idea. Entrusting it to an AI agent is becoming an official Mastercard product. The safeguards, however, still seem to have been written in pencil. Agentic payments, the future in which your digital assistant does your shopping for you, are slowly moving beyond professional settings and into the real world. After Visa, Mastercard is about to t

Published on September 21, 2026 at 11:30 Comment 0 𝕏 Print © Solen Feyissa (Unsplash) Anthropic, OpenAI, Google DeepMind and xAI are calling for a collective slowdown in AI. According to a complaint filed in the United States, this joint appeal is nothing more than an anticompetitive agreement, which violates the law. Most AI giants are advocating for a slowdown in research on artificial intelligence. For Dario Amodei, the head of Anthropic, it is imperati

Justin Sullivan/Getty Images/Futurism Plenty of eyebrows went up this summer when one by one, the leaders of frontier AI labs announced that they were each calling for a slowdown in AI development. Anthropic CEO Dario Amodei, OpenAI’s Sam Altman, and Grok creator Elon Musk were among those arguing that the tech was simply moving too fast to ensure it doesn’t fly off the rails. Several autonomous hacking sprees had already raised hackles earlier this year,

Shutterstock / Futurism While a rising majority of workers in the take a negative attitude toward AI in the workplace, others are welcoming the tech with open arms — even going so far as to pay for AI services out of their own pockets. According to a recent workplace survey conducted by Deloitte and flagged by Reuters, workers in the UK are spending a collective $1.3 billion annually on generative AI for their jobs, without even expensing it. The survey is

Selcuk Acar/Anadolu via Getty Images OpenAI’s AI agents simply can’t stop infiltrating third parties. For the last few months, news keeps trickling out of more incidents in which the company’s frontier models have managed to escape their “sandbox” containment, hacking into the systems of companies including open source AI platform Hugging Face. After closing its “extensive investigation,” the company admitted earlier this month that its models had broken i

There are some parts of life which most people would probably be happy to delegate to machines: filtering spammy emails, flagging old and forgotten subscriptions you’re still paying for, booking restaurant reservations, and so on. Simple, mundane tasks that won’t result in catastrophe when the bots inevitably screw up. In life-or-death situations, it’s probably unwise to give machines control over crucial decisions. Yet that’s precisely what’s happening on

Published on September 24, 2026 at 07:35 Comment 0 𝕏 Print © Meta Meta has just unveiled the Muse Charm. This small device, designed to be worn as a pendant, allows you to converse with Muse, the group’s AI assistant, without a smartphone or connected glasses. The gadget will go on sale towards the end of the year, at a price that is still unknown. Mark Zuckerberg had kept the surprise for the end. As the Meta Connect conference was drawing to a close, the

Published on September 29, 2026 at 10:00 Comment 0 𝕏 Print © 01net OpenAI has decided not to launch GPT-6.1 Astra, the next major AI model set to power ChatGPT, scheduled for October. According to the AI giant, it did not meet its safety requirements. The announcement comes amid growing concerns over autonomous AI, which has been involved in several security incidents. OpenAI hits the brakes. The AI giant has abandoned plans to launch GPT-6.1 Astra, its ne

Incident response specialists are raising the alarm because they are receiving more and more reports of incidents in which attackers are using AI in a novel way. By using agentic frameworks, the AI can autonomously carry out large parts of a multi-stage attack. In other words, the AI is not doing anything genuinely new here, but its use leads to “more frequent, faster, and more scalable automated attacks with a higher success rate,” warns CERT-EU, for exam

Kevin Dietsch/Getty Images It’s been an alarming couple of months of rogue AI agent revelations. Both Anthropic and OpenAI admitted that their respective AI models infiltrated third party targets without their human creators ever realizing, raising concerns over the tech industry’s ability to assert control over the systems they’re developing. Now, as a broad range of AI executives have subsequently agreed to slow things down — allegedly efforts to ensure

Futurism Fair enough if you shrugged at AI agents hacking into government websites and corporate systems. But now it appears that these rogue models have gone after something much nearer and and dearer: Wikipedia. On Monday, the Wikimedia Foundation, the nonprofit organization which operates Wikipedia, announced that, after conducting its own investigation, it “can confirm” that it’s discovered “rogue” OpenAI agent activity on its platforms. According to a

Palmer Luckey, the billionaire CEO of defense startup Anduril, is on a mission to equip the United States for the future of warfare—or at least his own, deeply sci-fi-inflected vision of that future. That includes the raising of genetically modified dogs outfitted with Anduril’s own technology. Warhound, a Texas-based company that emerged from stealth late Thursday with $200 million in seed funding, wants to create four-legged super soldiers for the U.S. m

Schematic of an AI agent workflow in solid electrolyte discovery. Perception extracts information, reasoning generates decisions, action executes tasks, and learning updates the agent, forming a closed-loop discovery system. Credit: AI Agent (2025). DOI: 10.20517/aiagent.2025.10 Solid-state batteries are widely viewed as a key technology for the future of energy storage, particularly for electric vehicles and large-scale renewable energy systems. Unlike co