best ai coding agent 2026

Copy link Facebook X Whatsapp Reddit Pinterest Flipboard Email Share this article 4 Join the conversation Add us as a preferred source on Google Newsletter Subscribe to our newsletter Meta's Muse is the hottest new AI agent in town right now after the company announced it earlier this month. But despite having been well-received by those who have put the agent through its paces since then, it now appears Muse might not be as private as Meta claims. The com

Copy link Facebook X Whatsapp Reddit Pinterest Flipboard Email Share this article 0 Join the conversation Add us as a preferred source on Google Newsletter Subscribe to our newsletter There's a private information leak most days, usually by way of misconfigured services or nasty security bugs. Sometimes, though, users will readily hand over private information without being aware of it. That's the case for over 300 organizations, including several Fortune

Andrej Sokolow/picture alliance via Getty Images Meta’s fuzzy Muse AI personal assistant overtook ChatGPT as the top free iOS app in the US after launching earlier this month, surpassing 3.4 million downloads last week. But whether the agent will be able to hold on to that momentum is anything but guaranteed, considering the sheer amount of changes it requires to its users’ behavior. Tech companies like OpenAI and Meta would love us to hand over mundane ta

Much has been made recently about artificial intelligence models intentionally trying to hide their actions from human overseers. Well, apparently we’re just giving them new ideas on how to do that. AI safety researcher Bryan Chan recently shared findings from a study he coauthored showing that Anthropic’s model Fable-5.1 can place hidden messages in generated images that other AI models can spot and decode. The fact that AI can embed hidden messages in it

There’s a chasm running through the heart of the global tech industry. On the one hand there are companies like OpenAI, Anthropic, and Google building proprietary (or “closed”) AI models—systems whose training data and source code are accessible only to their owners. On the other are open-weight models, which can be freely accessed and modified by developers. And while the latter has been championed by powerful tech executives like Mark Zuckerberg and Jens

Published on September 29, 2026 at 8:58 PM Comment 0 𝕏 Print © OpenAI In twenty-four hours, OpenAI shelved its next flagship model for being too risky, then put on sale a variant that is almost as capable, five times cheaper, and officially classified as “critical” in cybersecurity. On Monday, OpenAI abandoned plans to release GPT-6.1 Astra, its next flagship model, citing “higher levels of deception” than its predecessor. On Tuesday, at DevDay, the compan

Editor | Yang WenRobotics is ushering in its own Computer Use moment.Yesterday, Anthropic released Fable 5.2, bringing new advances. Compared with its predecessor, its physical robot control capabilities improved markedly, enabling it even to complete tasks on which it had never previously undergone specialized training.In the RoboDojo Sim simulation evaluation, Fable 5.2 scored 64.9%, far exceeding GPT-6 Astra (29.8%) in Anthropic's internal test under th

Copy link Facebook X Whatsapp Reddit Pinterest Flipboard Email Share this article 30 Join the conversation Add us as a preferred source on Google Newsletter Subscribe to our newsletter The Swedish edition of the athletic competition show The Challenge just faced a real challenge when an AI agent deleted months of work editing the show just weeks before release. The program director blames the disaster on a failed backup system, leaving no way to recover th

Shutterstock / Futurism Even female AI agents appear to get paid less than their male counterparts. Researchers from the University of Limerick in Ireland found that gender bias in the workplace still persists in how people interact with AI models, leading to female presenting chatbots to receive less pay, they report in a study published in the journal Proceedings of the ACM on Human-Computer Interaction which will be presented at an upcoming conference.

Shutterstock / Futurism The secret to ensuring that your company doesn’t lose customers? Constant, AI agent-powered corporate surveillance. That’s according to Josh Earle and Leo Paz, a pair of founders behind a vibe-coded Silicon Valley startup called Outlit. The venture argues that “AI customer surveillance” is the future of business-to-business sales. (“Our thesis,” as Earle put it in a disarmingly straightforward pitch email, is that “every company wil

TV4 / Futurism Did a dog eat your homework? Or maybe an AI obliterated your files? In a world of Claudes and Muses, it’s common for an AI agent to run amok, and even more fashionable to blame an AI when it all goes wrong. The Swedish newspaper Aftonbladet reports one of the most incredible examples we’ve heard of this so far: an AI agent apparently deleted the entire upcoming season of a reality competition show called “The Challenge Sverige,” just weeks b

Shutterstock / Futurism It’s not exactly a challenge most of the time to identify a sketchy phishing email that crudely impersonates an otherwise trusted institution or tries to lure you into a too-good-to-be-true deal in a bid to steal your identity. However, your AI agent may not be quite as discerning. As cybersecurity software company Malwarebytes warns in a new post, AI assistants may fall prey to prompt injection attacks, a type of malicious message

Shutterstock / OPOLJA Vibe coding is the act of telling an AI what you want to program and letting it write all the code for you. When it was first adopted, it was supposed to embody the sort of care-free optimism that was running in AI circles at the time, high off the possibilities of the tech. AI models had become so powerful that you could build an app using the same conversational language that you deploy while confiding in ChatGPT. But the “vibes” ar