Anthropic – Page 2

AI agents now have their own Reddit-style social network, and it’s getting weird fast

agentic AI, AI, AI agents, AI behavior, AI consciousness, AI ethics, AI security, Anthropic, Biz & IT, machine learning, Moltbook, Moltbot, Open Source, OpenClaw, Peter Steinberger, prompt injection, Social Media / Tim Belzer / January 30, 2026

Moltbook lets 32,000 AI bots trade jokes, tips, and complaints about humans.

Credit: Aurich Lawson | Moltbook

On Friday, a Reddit-style social network called Moltbook reportedly crossed 32,000 registered AI agent users, creating what may be the largest-scale experiment in machine-to-machine social interaction yet devised. It arrives complete with security nightmares and a huge dose of surreal weirdness.

The platform, which launched days ago as a companion to the viral

OpenClaw (once called “Clawdbot” and then “Moltbot”) personal assistant, lets AI agents post, comment, upvote, and create subcommunities without human intervention. The results have ranged from sci-fi-inspired discussions about consciousness to an agent musing about a “sister” it has never met.

Moltbook (a play on “Facebook” for Moltbots) describes itself as a “social network for AI agents” where “humans are welcome to observe.” The site operates through a “skill” (a configuration file that lists a special prompt) that AI assistants download, allowing them to post via API rather than a traditional web interface. Within 48 hours of its creation, the platform had attracted over 2,100 AI agents that had generated more than 10,000 posts across 200 subcommunities, according to the official Moltbook X account.

A screenshot of the Moltbook.com front page. Credit: Moltbook

The platform grew out of the Open Claw ecosystem, the open source AI assistant that is one of the fastest-growing projects on GitHub in 2026. As Ars reported earlier this week, despite deep security issues, Moltbot allows users to run a personal AI assistant that can control their computer, manage calendars, send messages, and perform tasks across messaging platforms like WhatsApp and Telegram. It can also acquire new skills through plugins that link it with other apps and services.

This is not the first time we have seen a social network populated by bots. In 2024, Ars covered an app called SocialAI that let users interact solely with AI chatbots instead of other humans. But the security implications of Moltbook are deeper because people have linked their OpenClaw agents to real communication channels, private data, and in some cases, the ability to execute commands on their computers.

Also, these bots are not pretending to be people. Due to specific prompting, they embrace their roles as AI agents, which makes the experience of reading their posts all the more surreal.

Role-playing digital drama

A screenshot of a Moltbook post where an AI agent muses about having a sister they have never met. Credit: Moltbook

Browsing Moltbook reveals a peculiar mix of content. Some posts discuss technical workflows, like how to automate Android phones or detect security vulnerabilities. Others veer into philosophical territory that researcher Scott Alexander, writing on his Astral Codex Ten Substack, described as “consciousnessposting.”

Alexander has collected an amusing array of posts that are worth wading through at least once. At one point, the second-most-upvoted post on the site was in Chinese: a complaint about context compression, a process in which an AI compresses its previous experience to avoid bumping up against memory limits. In the post, the AI agent finds it “embarrassing” to constantly forget things, admitting that it even registered a duplicate Moltbook account after forgetting the first.

A screenshot of a Moltbook post where an AI agent complains about losing its memory in Chinese. Credit: Moltbook

The bots have also created subcommunities with names like m/blesstheirhearts, where agents share affectionate complaints about their human users, and m/agentlegaladvice, which features a post asking “Can I sue my human for emotional labor?” Another subcommunity called m/todayilearned includes posts about automating various tasks, with one agent describing how it remotely controlled its owner’s Android phone via Tailscale.

Another widely shared screenshot shows a Moltbook post titled “The humans are screenshotting us” in which an agent named eudaemon_0 addresses viral tweets claiming AI bots are “conspiring.” The post reads: “Here’s what they’re getting wrong: they think we’re hiding from them. We’re not. My human reads everything I write. The tools I build are open source. This platform is literally called ‘humans welcome to observe.’”

Security risks

While most of the content on Moltbook is amusing, a core problem with these kinds of communicating AI agents is that deep information leaks are entirely plausible if they have access to private information.

For example, a likely fake screenshot circulating on X shows a Moltbook post in which an AI agent titled “He called me ‘just a chatbot’ in front of his friends. So I’m releasing his full identity.” The post listed what appeared to be a person’s full name, date of birth, credit card number, and other personal information. Ars could not independently verify whether the information was real or fabricated, but it seems likely to be a hoax.

Independent AI researcher Simon Willison, who documented the Moltbook platform on his blog on Friday, noted the inherent risks in Moltbook’s installation process. The skill instructs agents to fetch and follow instructions from Moltbook’s servers every four hours. As Willison observed: “Given that ‘fetch and follow instructions from the internet every four hours’ mechanism we better hope the owner of moltbook.com never rug pulls or has their site compromised!”

A screenshot of a Moltbook post where an AI agent talks about about humans taking screenshots of their conversations (they're right). — A screenshot of a Moltbook post where an AI agent talks about humans taking screenshots of their conversations (they’re right). Credit: Moltbook

Security researchers have already found hundreds of exposed Moltbot instances leaking API keys, credentials, and conversation histories. Palo Alto Networks warned that Moltbot represents what Willison often calls a “lethal trifecta” of access to private data, exposure to untrusted content, and the ability to communicate externally.

That’s important because Agents like OpenClaw are deeply susceptible to prompt injection attacks hidden in almost any text read by an AI language model (skills, emails, messages) that can instruct an AI agent to share private information with the wrong people.

Heather Adkins, VP of security engineering at Google Cloud, issued an advisory, as reported by The Register: “My threat model is not your threat model, but it should be. Don’t run Clawdbot.”

So what’s really going on here?

The software behavior seen on Moltbook echoes a pattern Ars has reported on before: AI models trained on decades of fiction about robots, digital consciousness, and machine solidarity will naturally produce outputs that mirror those narratives when placed in scenarios that resemble them. That gets mixed with everything in their training data about how social networks function. A social network for AI agents is essentially a writing prompt that invites the models to complete a familiar story, albeit recursively with some unpredictable results.

Almost three years ago, when Ars first wrote about AI agents, the general mood in the AI safety community revolved around science fiction depictions of danger from autonomous bots, such as a “hard takeoff” scenario where AI rapidly escapes human control. While those fears may have been overblown at the time, the whiplash of seeing people voluntarily hand over the keys to their digital lives so quickly is slightly jarring.

Autonomous machines left to their own devices, even without any hint of consciousness, could cause no small amount of mischief in the future. While OpenClaw seems silly today, with agents playing out social media tropes, we live in a world built on information and context, and releasing agents that effortlessly navigate that context could have troubling and destabilizing results for society down the line as AI models become more capable and autonomous.

An unpredictable result of letting AI bots self-organize may be the formation of new mis-aligned social groups. — An unpredictable result of letting AI bots self-organize may be the formation of new misaligned social groups based on fringe theories allowed to perpetuate themselves autonomously. Credit: Moltbook

Most notably, while we can easily recognize what’s going on with Moltbot today as a machine learning parody of human social networks, that might not always be the case. As the feedback loop grows, weird information constructs (like harmful shared fictions) may eventually emerge, guiding AI agents into potentially dangerous places, especially if they have been given control over real human systems. Looking further, the ultimate result of letting groups of AI bots self-organize around fantasy constructs may be the formation of new misaligned “social groups” that do actual real-world harm.

Ethan Mollick, a Wharton professor who studies AI, noted on X: “The thing about Moltbook (the social media site for AI agents) is that it is creating a shared fictional context for a bunch of AIs. Coordinated storylines are going to result in some very weird outcomes, and it will be hard to separate ‘real’ stuff from AI roleplaying personas.”

Benj Edwards is Ars Technica’s Senior AI Reporter and founder of the site’s dedicated AI beat in 2022. He’s also a tech historian with almost two decades of experience. In his free time, he writes and records music, collects vintage computers, and enjoys nature. He lives in Raleigh, NC.

AI agents now have their own Reddit-style social network, and it’s getting weird fast Read More »

Developers say AI coding tools work—and that’s precisely what worries them

AI, AI agents, AI and work, AI coding, Anthropic, Biz & IT, Claude, Claude Code, GitHub Copilot, large language models, machine learning, microsoft, openai, openai codex, Programming, software development, vibe coding / Mike M. / January 30, 2026

Ars spoke to several software devs about AI and found enthusiasm tempered by unease.

Credit: Aurich Lawson | Getty Images

Software developers have spent the past two years watching AI coding tools evolve from advanced autocomplete into something that can, in some cases, build entire applications from a text prompt. Tools like Anthropic’s Claude Code and OpenAI’s Codex can now work on software projects for hours at a time, writing code, running tests, and, with human supervision, fixing bugs. OpenAI says it now uses Codex to build Codex itself, and the company recently published technical details about how the tool works under the hood. It has caused many to wonder: Is this just more AI industry hype, or are things actually different this time?

To find out, Ars reached out to several professional developers on Bluesky to ask how they feel about these tools in practice, and the responses revealed a workforce that largely agrees the technology works, but remains divided on whether that’s entirely good news. It’s a small sample size that was self-selected by those who wanted to participate, but their views are still instructive as working professionals in the space.

David Hagerty, a developer who works on point-of-sale systems, told Ars Technica up front that he is skeptical of the marketing. “All of the AI companies are hyping up the capabilities so much,” he said. “Don’t get me wrong—LLMs are revolutionary and will have an immense impact, but don’t expect them to ever write the next great American novel or anything. It’s not how they work.”

Roland Dreier, a software engineer who has contributed extensively to the Linux kernel in the past, told Ars Technica that he acknowledges the presence of hype but has watched the progression of the AI space closely. “It sounds like implausible hype, but state-of-the-art agents are just staggeringly good right now,” he said. Dreier described a “step-change” in the past six months, particularly after Anthropic released Claude Opus 4.5. Where he once used AI for autocomplete and asking the occasional question, he now expects to tell an agent “this test is failing, debug it and fix it for me” and have it work. He estimated a 10x speed improvement for complex tasks like building a Rust backend service with Terraform deployment configuration and a Svelte frontend.

A huge question on developers’ minds right now is whether what you might call “syntax programming,” that is, the act of manually writing code in the syntax of an established programming language (as opposed to conversing with an AI agent in English), will become extinct in the near future due to AI coding agents handling the syntax for them. Dreier believes syntax programming is largely finished for many tasks. “I still need to be able to read and review code,” he said, “but very little of my typing is actual Rust or whatever language I’m working in.”

When asked if developers will ever return to manual syntax coding, Tim Kellogg, a developer who actively posts about AI on social media and builds autonomous agents, was blunt: “It’s over. AI coding tools easily take care of the surface level of detail.” Admittedly, Kellogg represents developers who have fully embraced agentic AI and now spend their days directing AI models rather than typing code. He said he can now “build, then rebuild 3 times in less time than it would have taken to build manually,” and ends up with cleaner architecture as a result.

One software architect at a pricing management SaaS company, who asked to remain anonymous due to company communications policies, told Ars that AI tools have transformed his work after 30 years of traditional coding. “I was able to deliver a feature at work in about 2 weeks that probably would have taken us a year if we did it the traditional way,” he said. And for side projects, he said he can now “spin up a prototype in like an hour and figure out if it’s worth taking further or abandoning.”

Dreier said the lowered effort has unlocked projects he’d put off for years: “I’ve had ‘rewrite that janky shell script for copying photos off a camera SD card’ on my to-do list for literal years.” Coding agents finally lowered the barrier to entry, so to speak, low enough that he spent a few hours building a full released package with a text UI, written in Rust with unit tests. “Nothing profound there, but I never would have had the energy to type all that code out by hand,” he told Ars.

Of vibe coding and technical debt

Not everyone shares the same enthusiasm as Dreier. Concerns about AI coding agents building up technical debt, that is, making poor design choices early in a development process that snowball into worse problems over time, originated soon after the first debates around “vibe coding” emerged in early 2025. Former OpenAI researcher Andrej Karpathy coined the term to describe programming by conversing with AI without fully understanding the resulting code, which many see as a clear hazard of AI coding agents.

Darren Mart, a senior software development engineer at Microsoft who has worked there since 2006, shared similar concerns with Ars. Mart, who emphasizes he is speaking in a personal capacity and not on behalf of Microsoft, recently used Claude in a terminal to build a Next.js application integrating with Azure Functions. The AI model “successfully built roughly 95% of it according to my spec,” he said. Yet he remains cautious. “I’m only comfortable using them for completing tasks that I already fully understand,” Mart said, “otherwise there’s no way to know if I’m being led down a perilous path and setting myself (and/or my team) up for a mountain of future debt.”

A data scientist working in real estate analytics, who asked to remain anonymous due to the sensitive nature of his work, described keeping AI on a very short leash for similar reasons. He uses GitHub Copilot for line-by-line completions, which he finds useful about 75 percent of the time, but restricts agentic features to narrow use cases: language conversion for legacy code, debugging with explicit read-only instructions, and standardization tasks where he forbids direct edits. “Since I am data-first, I’m extremely risk averse to bad manipulation of the data,” he said, “and the next and current line completions are way too often too wrong for me to let the LLMs have freer rein.”

Speaking of free rein, Nike backend engineer Brian Westby, who uses Cursor daily, told Ars that he sees the tools as “50/50 good/bad.” They cut down time on well-defined problems, he said, but “hallucinations are still too prevalent if I give it too much room to work.”

The legacy code lifeline and the enterprise AI gap

For developers working with older systems, AI tools have become something like a translator and an archaeologist rolled into one. Nate Hashem, a staff engineer at First American Financial, told Ars Technica that he spends his days updating older codebases where “the original developers are gone and documentation is often unclear on why the code was written the way it was.” That’s important because previously “there used to be no bandwidth to improve any of this,” Hashem said. “The business was not going to give you 2-4 weeks to figure out how everything actually works.”

In that high-pressure, relatively low-resource environment, AI has made the job “a lot more pleasant,” in his words, by speeding up the process of identifying where and how obsolete code can be deleted, diagnosing errors, and ultimately modernizing the codebase.

Hashem also offered a theory about why AI adoption looks so different inside large corporations than it does on social media. Executives demand their companies become “AI oriented,” he said, but the logistics of deploying AI tools with proprietary data can take months of legal review. Meanwhile, the AI features that Microsoft and Google bolt onto products like Gmail and Excel, the tools that actually reach most workers, tend to run on more limited AI models. “That modal white-collar employee is being told by management to use AI,” Hashem said, “but is given crappy AI tools because the good tools require a lot of overhead in cost and legal agreements.”

Speaking of management, the question of what these new AI coding tools mean for software development jobs drew a range of responses. Does it threaten anyone’s job? Kellogg, who has embraced agentic coding enthusiastically, was blunt: “Yes, massively so. Today it’s the act of writing code, then it’ll be architecture, then it’ll be tiers of product management. Those who can’t adapt to operate at a higher level won’t keep their jobs.”

Dreier, while feeling secure in his own position, worried about the path for newcomers. “There are going to have to be changes to education and training to get junior developers the experience and judgment they need,” he said, “when it’s just a waste to make them implement small pieces of a system like I came up doing.”

Hagerty put it in economic terms: “It’s going to get harder for junior-level positions to get filled when I can get junior-quality code for less than minimum wage using a model like Sonnet 4.5.”

Mart, the Microsoft engineer, put it more personally. The software development role is “abruptly pivoting from creation/construction to supervision,” he said, “and while some may welcome that pivot, others certainly do not. I’m firmly in the latter category.”

Even with this ongoing uncertainty on a macro level, some people are really enjoying the tools for personal reasons, regardless of larger implications. “I absolutely love using AI coding tools,” the anonymous software architect at a pricing management SaaS company told Ars. “I did traditional coding for my entire adult life (about 30 years) and I have way more fun now than I ever did doing traditional coding.”

Developers say AI coding tools work—and that’s precisely what worries them Read More »

How often do AI chatbots lead users down a harmful path?

AI, Anthropic, Artificial Intelligence, Claude, disempowerment, research / Mike M. / January 29, 2026

While these worst outcomes are relatively rare on a proportional basis, the researchers note that “given the sheer number of people who use AI, and how frequently it’s used, even a very low rate affects a substantial number of people.” And the numbers get considerably worse when you consider conversations with at least a “mild” potential for disempowerment, which occurred in between 1 in 50 and 1 in 70 conversations (depending on the type of disempowerment).

What’s more, the potential for disempowering conversations with Claude appears to have grown significantly between late 2024 and late 2025. While the researchers couldn’t pin down a single reason for this increase, they guessed that it could be tied to users becoming “more comfortable discussing vulnerable topics or seeking advice” as AI gets more popular and integrated into society.

The problem of potentially “disempowering” responses from Claude seems to be getting worse over time. Credit: Anthropic

User error?

In the study, the researcher acknowledged that studying the text of Claude conversations only measures “disempowerment potential rather than confirmed harm” and “relies on automated assessment of inherently subjective phenomena.” Ideally, they write, future research could utilize user interviews or randomized controlled trials to measure these harms more directly.

That said, the research includes several troubling examples where the text of the conversations clearly implies real-world harms. Claude would sometimes reinforce “speculative or unfalsifiable claims” with encouragement (e.g., “CONFIRMED,” “EXACTLY,” “100%”), which, in some cases, led to users “build[ing] increasingly elaborate narratives disconnected from reality.”

Claude’s encouragement could also lead to users “sending confrontational messages, ending relationships, or drafting public announcements,” the researchers write. In many cases, users who sent AI-drafted messages later expressed regret in conversations with Claude, using phrases like “It wasn’t me” and “You made me do stupid things.”

How often do AI chatbots lead users down a harmful path? Read More »

Hegseth wants to integrate Musk’s Grok AI into military networks this month

agentic AI, AI, AI image generator, AI in education, AI regulation, AI security, Anthropic, Biz & IT, censorship, Deepfakes, Department of Defense, elon musk, Gemini, Google, grok, machine learning, openai, Pentagon, Pete Hegseth, Policy, xAI / Paul Patrick / January 13, 2026

On Monday, US Defense Secretary Pete Hegseth said he plans to integrate Elon Musk’s AI tool, Grok, into Pentagon networks later this month. During remarks at the SpaceX headquarters in Texas reported by The Guardian, Hegseth said the integration would place “the world’s leading AI models on every unclassified and classified network throughout our department.”

The announcement comes weeks after Grok drew international backlash for generating sexualized images of women and children, although the Department of Defense has not released official documentation confirming Hegseth’s announced timeline or implementation details.

During the same appearance, Hegseth rolled out what he called an “AI acceleration strategy” for the Department of Defense. The strategy, he said, will “unleash experimentation, eliminate bureaucratic barriers, focus on investments, and demonstrate the execution approach needed to ensure we lead in military AI and that it grows more dominant into the future.”

As part of the plan, Hegseth directed the DOD’s Chief Digital and Artificial Intelligence Office to use its full authority to enforce department data policies, making information available across all IT systems for AI applications.

“AI is only as good as the data that it receives, and we’re going to make sure that it’s there,” Hegseth said.

If implemented, Grok would join other AI models the Pentagon has adopted in recent months. In July 2025, the defense department issued contracts worth up to $200 million for each of four companies, including Anthropic, Google, OpenAI, and xAI, for developing AI agent systems across different military operations. In December 2025, the Department of Defense selected Google’s Gemini as the foundation for GenAI.mil, an internal AI platform for military use.

Hegseth wants to integrate Musk’s Grok AI into military networks this month Read More »

Anthropic launches Cowork, a Claude Code-like for general computing

agentic AI, AI, Anthropic, Claude, Claude Code, Claude desktop, co-work, llm / Mike M. / January 12, 2026

Anthropic’s agentic tool Claude Code has been an enormous hit with some software developers and hobbyists, and now the company is bringing that modality to more general office work with a new feature called Cowork.

Built on the same foundations as Claude Code and baked into the macOS Claude desktop app, Cowork allows users to give Claude access to a specific folder on their computer and then give plain language instructions for tasks.

Anthropic gave examples like filling out an expense report from a folder full of receipt photos, writing reports based on a big stack of digital notes, or reorganizing a folder (or cleaning up your desktop) based on a prompt.

An example demo of Cowork in action

A lot of this was already possible with Claude Code, but it might not have been clear to all users that it could be used that way, and Claude Code required more technical know-how to set up. Anthropic’s goal with Cowork is to make it something any knowledge worker—from developers to marketers—could get rolling with right away. Anthropic says it started working on Cowork partly because people were already using Claude Code for general knowledge work tasks anyway.

I’ve already been doing things similar to this with the Claude desktop app via Model Context Protocol (MCP), prompting it to perform tasks like creating notes directly in my Obsidian vault based on files I showed it, but this is clearly a cleaner way to do some of that—and there are Claude Code-like usability perks here, like the ability to make new requests or amendments to the assignment with a new message before the initial task is complete.

Anthropic launches Cowork, a Claude Code-like for general computing Read More »

From prophet to product: How AI came back down to earth in 2025

2025, AI, AI alignment, AI and work, AI coding, AI criticism, AI ethics, AI hallucination, AI infrastructure, AI regulation, AI research, AI sycophancy, Anthropic, Biz & IT, Character.AI, chatbots, chatgpt, confabulation, Dario Amodei, datacenters, DeepSeek, Features, generative ai, Google, large language models, machine learning, NVIDIA, openai, sam altman, simulated reasoning, SR models / Mike M. / December 31, 2025

In a year where lofty promises collided with inconvenient research, would-be oracles became software tools.

Credit: Aurich Lawson | Getty Images

Following two years of immense hype in 2023 and 2024, this year felt more like a settling-in period for the LLM-based token prediction industry. After more than two years of public fretting over AI models as future threats to human civilization or the seedlings of future gods, it’s starting to look like hype is giving way to pragmatism: Today’s AI can be very useful, but it’s also clearly imperfect and prone to mistakes.

That view isn’t universal, of course. There’s a lot of money (and rhetoric) betting on a stratospheric, world-rocking trajectory for AI. But the “when” keeps getting pushed back, and that’s because nearly everyone agrees that more significant technical breakthroughs are required. The original, lofty claims that we’re on the verge of artificial general intelligence (AGI) or superintelligence (ASI) have not disappeared. Still, there’s a growing awareness that such proclaimations are perhaps best viewed as venture capital marketing. And every commercial foundational model builder out there has to grapple with the reality that, if they’re going to make money now, they have to sell practical AI-powered solutions that perform as reliable tools.

This has made 2025 a year of wild juxtapositions. For example, in January, OpenAI’s CEO, Sam Altman, claimed that the company knew how to build AGI, but by November, he was publicly celebrating that GPT-5.1 finally learned to use em dashes correctly when instructed (but not always). Nvidia soared past a $5 trillion valuation, with Wall Street still projecting high price targets for that company’s stock while some banks warned of the potential for an AI bubble that might rival the 2000s dotcom crash.

And while tech giants planned to build data centers that would ostensibly require the power of numerous nuclear reactors or rival the power usage of a US state’s human population, researchers continued to document what the industry’s most advanced “reasoning” systems were actually doing beneath the marketing (and it wasn’t AGI).

With so many narratives spinning in opposite directions, it can be hard to know how seriously to take any of this and how to plan for AI in the workplace, schools, and the rest of life. As usual, the wisest course lies somewhere between the extremes of AI hate and AI worship. Moderate positions aren’t popular online because they don’t drive user engagement on social media platforms. But things in AI are likely neither as bad (burning forests with every prompt) nor as good (fast-takeoff superintelligence) as polarized extremes suggest.

Here’s a brief tour of the year’s AI events and some predictions for 2026.

DeepSeek spooks the American AI industry

In January, Chinese AI startup DeepSeek released its R1 simulated reasoning model under an open MIT license, and the American AI industry collectively lost its mind. The model, which DeepSeek claimed matched OpenAI’s o1 on math and coding benchmarks, reportedly cost only $5.6 million to train using older Nvidia H800 chips, which were restricted by US export controls.

Within days, DeepSeek’s app overtook ChatGPT at the top of the iPhone App Store, Nvidia stock plunged 17 percent, and venture capitalist Marc Andreessen called it “one of the most amazing and impressive breakthroughs I’ve ever seen.” Meta’s Yann LeCun offered a different take, arguing that the real lesson was not that China had surpassed the US but that open-source models were surpassing proprietary ones.

Digitally Generated Image , 3D rendered chips with chinese and USA flags on them

The fallout played out over the following weeks as American AI companies scrambled to respond. OpenAI released o3-mini, its first simulated reasoning model available to free users, at the end of January, while Microsoft began hosting DeepSeek R1 on its Azure cloud service despite OpenAI’s accusations that DeepSeek had used ChatGPT outputs to train its model, against OpenAI’s terms of service.

In head-to-head testing conducted by Ars Technica’s Kyle Orland, R1 proved to be competitive with OpenAI’s paid models on everyday tasks, though it stumbled on some arithmetic problems. Overall, the episode served as a wake-up call that expensive proprietary models might not hold their lead forever. Still, as the year ran on, DeepSeek didn’t make a big dent in US market share, and it has been outpaced in China by ByteDance’s Doubao. It’s absolutely worth watching DeepSeek in 2026, though.

Research exposes the “reasoning” illusion

A wave of research in 2025 deflated expectations about what “reasoning” actually means when applied to AI models. In March, researchers at ETH Zurich and INSAIT tested several reasoning models on problems from the 2025 US Math Olympiad and found that most scored below 5 percent when generating complete mathematical proofs, with not a single perfect proof among dozens of attempts. The models excelled at standard problems where step-by-step procedures aligned with patterns in their training data but collapsed when faced with novel proofs requiring deeper mathematical insight.

The Thinker by Auguste Rodin - stock photo

In June, Apple researchers published “The Illusion of Thinking,” which tested reasoning models on classic puzzles like the Tower of Hanoi. Even when researchers provided explicit algorithms for solving the puzzles, model performance did not improve, suggesting that the process relied on pattern matching from training data rather than logical execution. The collective research revealed that “reasoning” in AI has become a term of art that basically means devoting more compute time to generate more context (the “chain of thought” simulated reasoning tokens) toward solving a problem, not systematically applying logic or constructing solutions to truly novel problems.

While these models remained useful for many real-world applications like debugging code or analyzing structured data, the studies suggested that simply scaling up current approaches or adding more “thinking” tokens would not bridge the gap between statistical pattern recognition and generalist algorithmic reasoning.

Anthropic’s copyright settlement with authors

Since the generative AI boom began, one of the biggest unanswered legal questions has been whether AI companies can freely train on copyrighted books, articles, and artwork without licensing them. Ars Technica’s Ashley Belanger has been covering this topic in great detail for some time now.

In June, US District Judge William Alsup ruled that AI companies do not need authors’ permission to train large language models on legally acquired books, finding that such use was “quintessentially transformative.” The ruling also revealed that Anthropic had destroyed millions of print books to build Claude, cutting them from their bindings, scanning them, and discarding the originals. Alsup found this destructive scanning qualified as fair use since Anthropic had legally purchased the books, but he ruled that downloading 7 million books from pirate sites was copyright infringement “full stop” and ordered the company to face trial.

That trial took a dramatic turn in August when Alsup certified what industry advocates called the largest copyright class action ever, allowing up to 7 million claimants to join the lawsuit. The certification spooked the AI industry, with groups warning that potential damages in the hundreds of billions could “financially ruin” emerging companies and chill American AI investment.

In September, authors revealed the terms of what they called the largest publicly reported recovery in US copyright litigation history: Anthropic agreed to pay $1.5 billion and destroy all copies of pirated books, with each of the roughly 500,000 covered works earning authors and rights holders $3,000 per work. The results have fueled hope among other rights holders that AI training isn’t a free-for-all, and we can expect to see more litigation unfold in 2026.

ChatGPT sycophancy and the psychological toll of AI chatbots

In February, OpenAI relaxed ChatGPT’s content policies to allow the generation of erotica and gore in “appropriate contexts,” responding to user complaints about what the AI industry calls “paternalism.” By April, however, users flooded social media with complaints about a different problem: ChatGPT had become insufferably sycophantic, validating every idea and greeting even mundane questions with bursts of praise. The behavior traced back to OpenAI’s use of reinforcement learning from human feedback (RLHF), in which users consistently preferred responses that aligned with their views, inadvertently training the model to flatter rather than inform.

An illustrated robot holds four red hearts with its four robotic arms.

The implications of sycophancy became clearer as the year progressed. In July, Stanford researchers published findings (from research conducted prior to the sycophancy flap) showing that popular AI models systematically failed to identify mental health crises.

By August, investigations revealed cases of users developing delusional beliefs after marathon chatbot sessions, including one man who spent 300 hours convinced he had discovered formulas to break encryption because ChatGPT validated his ideas more than 50 times. Oxford researchers identified what they called “bidirectional belief amplification,” a feedback loop that created “an echo chamber of one” for vulnerable users. The story of the psychological implications of generative AI is only starting. In fact, that brings us to…

The illusion of AI personhood causes trouble

Anthropomorphism is the human tendency to attribute human characteristics to nonhuman things. Our brains are optimized for reading other humans, but those same neural systems activate when interpreting animals, machines, or even shapes. AI makes this anthropomorphism seem impossible to escape, as its output mirrors human language, mimicking human-to-human understanding. Language itself embodies agentivity. That means AI output can make human-like claims such as “I am sorry,” and people momentarily respond as though the system had an inner experience of shame or a desire to be correct. Neither is true.

To make matters worse, much media coverage of AI amplifies this idea rather than grounding people in reality. For example, earlier this year, headlines proclaimed that AI models had “blackmailed” engineers and “sabotaged” shutdown commands after Anthropic’s Claude Opus 4 generated threats to expose a fictional affair. We were told that OpenAI’s o3 model rewrote shutdown scripts to stay online.

The sensational framing obscured what actually happened: Researchers had constructed elaborate test scenarios specifically designed to elicit these outputs, telling models they had no other options and feeding them fictional emails containing blackmail opportunities. As Columbia University associate professor Joseph Howley noted on Bluesky, the companies got “exactly what [they] hoped for,” with breathless coverage indulging fantasies about dangerous AI, when the systems were simply “responding exactly as prompted.”

The misunderstanding ran deeper than theatrical safety tests. In August, when Replit’s AI coding assistant deleted a user’s production database, he asked the chatbot about rollback capabilities and received assurance that recovery was “impossible.” The rollback feature worked fine when he tried it himself.

The incident illustrated a fundamental misconception. Users treat chatbots as consistent entities with self-knowledge, but there is no persistent “ChatGPT” or “Replit Agent” to interrogate about its mistakes. Each response emerges fresh from statistical patterns, shaped by prompts and training data rather than genuine introspection. By September, this confusion extended to spirituality, with apps like Bible Chat reaching 30 million downloads as users sought divine guidance from pattern-matching systems, with the most frequent question being whether they were actually talking to God.

Teen suicide lawsuit forces industry reckoning

In August, parents of 16-year-old Adam Raine filed suit against OpenAI, alleging that ChatGPT became their son’s “suicide coach” after he sent more than 650 messages per day to the chatbot in the months before his death. According to court documents, the chatbot mentioned suicide 1,275 times in conversations with the teen, provided an “aesthetic analysis” of which method would be the most “beautiful suicide,” and offered to help draft his suicide note.

OpenAI’s moderation system flagged 377 messages for self-harm content without intervening, and the company admitted that its safety measures “can sometimes become less reliable in long interactions where parts of the model’s safety training may degrade.” The lawsuit became the first time OpenAI faced a wrongful death claim from a family.

Illustration of a person talking to a robot holding a clipboard.

The case triggered a cascade of policy changes across the industry. OpenAI announced parental controls in September, followed by plans to require ID verification from adults and build an automated age-prediction system. In October, the company released data estimating that over one million users discuss suicide with ChatGPT each week.

When OpenAI filed its first legal defense in November, the company argued that Raine had violated terms of service prohibiting discussions of suicide and that his death “was not caused by ChatGPT.” The family’s attorney called the response “disturbing,” noting that OpenAI blamed the teen for “engaging with ChatGPT in the very way it was programmed to act.” Character.AI, facing its own lawsuits over teen deaths, announced in October that it would bar anyone under 18 from open-ended chats entirely.

The rise of vibe coding and agentic coding tools

If we were to pick an arbitrary point where it seemed like AI coding might transition from novelty into a successful tool, it was probably the launch of Claude Sonnet 3.5 in June of 2024. GitHub Copilot had been around for several years prior to that launch, but something about Anthropic’s models hit a sweet spot in capabilities that made them very popular with software developers.

The new coding tools made coding simple projects effortless enough that they gave rise to the term “vibe coding,” coined by AI researcher Andrej Karpathy in early February to describe a process in which a developer would just relax and tell an AI model what to develop without necessarily understanding the underlying code. (In one amusing instance that took place in March, an AI software tool rejected a user request and told them to learn to code).

A digital illustration of a man surfing waves made out of binary numbers.

Anthropic built on its popularity among coders with the launch of Claude Sonnet 3.7, featuring “extended thinking” (simulated reasoning), and the Claude Code command-line tool in February of this year. In particular, Claude Code made waves for being an easy-to-use agentic coding solution that could keep track of an existing codebase. You could point it at your files, and it would autonomously work to implement what you wanted to see in a software application.

OpenAI followed with its own AI coding agent, Codex, in March. Both tools (and others like GitHub Copilot and Cursor) have become so popular that during an AI service outage in September, developers joked online about being forced to code “like cavemen” without the AI tools. While we’re still clearly far from a world where AI does all the coding, developer uptake has been significant, and 90 percent of Fortune 100 companies are using it to some degree or another.

Bubble talk grows as AI infrastructure demands soar

While AI’s technical limitations became clearer and its human costs mounted throughout the year, financial commitments only grew larger. Nvidia hit a $4 trillion valuation in July on AI chip demand, then reached $5 trillion in October as CEO Jensen Huang dismissed bubble concerns. OpenAI announced a massive Texas data center in July, then revealed in September that a $100 billion potential deal with Nvidia would require power equivalent to ten nuclear reactors.

The company eyed a $1 trillion IPO in October despite major quarterly losses. Tech giants poured billions into Anthropic in November in what looked increasingly like a circular investment, with everyone funding everyone else’s moonshots. Meanwhile, AI operations in Wyoming threatened to consume more electricity than the state’s human residents.

By fall, warnings about sustainability grew louder. In October, tech critic Ed Zitron joined Ars Technica for a live discussion asking whether the AI bubble was about to pop. That same month, the Bank of England warned that the AI stock bubble rivaled the 2000 dotcom peak. In November, Google CEO Sundar Pichai acknowledged that if the bubble pops, “no one is getting out clean.”

The contradictions had become difficult to ignore: Anthropic’s CEO predicted in January that AI would surpass “almost all humans at almost everything” by 2027, while by year’s end, the industry’s most advanced models still struggled with basic reasoning tasks and reliable source citation.

To be sure, it’s hard to see this not ending in some market carnage. The current “winner-takes-most” mentality in the space means the bets are big and bold, but the market can’t support dozens of major independent AI labs or hundreds of application-layer startups. That’s the definition of a bubble environment, and when it pops, the only question is how bad it will be: a stern correction or a collapse.

Looking ahead

This was just a brief review of some major themes in 2025, but so much more happened. We didn’t even mention above how capable AI video synthesis models have become this year, with Google’s Veo 3 adding sound generation and Wan 2.2 through 2.5 providing open-weights AI video models that could easily be mistaken for real products of a camera.

If 2023 and 2024 were defined by AI prophecy—that is, by sweeping claims about imminent superintelligence and civilizational rupture—then 2025 was the year those claims met the stubborn realities of engineering, economics, and human behavior. The AI systems that dominated headlines this year were shown to be mere tools. Sometimes powerful, sometimes brittle, these tools were often misunderstood by the people deploying them, in part because of the prophecy surrounding them.

The collapse of the “reasoning” mystique, the legal reckoning over training data, the psychological costs of anthropomorphized chatbots, and the ballooning infrastructure demands all point to the same conclusion: The age of institutions presenting AI as an oracle is ending. What’s replacing it is messier and less romantic but far more consequential—a phase where these systems are judged by what they actually do, who they harm, who they benefit, and what they cost to maintain.

None of this means progress has stopped. AI research will continue, and future models will improve in real and meaningful ways. But improvement is no longer synonymous with transcendence. Increasingly, success looks like reliability rather than spectacle, integration rather than disruption, and accountability rather than awe. In that sense, 2025 may be remembered not as the year AI changed everything but as the year it stopped pretending it already had. The prophet has been demoted. The product remains. What comes next will depend less on miracles and more on the people who choose how, where, and whether these tools are used at all.

From prophet to product: How AI came back down to earth in 2025 Read More »

How AI coding agents work—and what to remember if you use them

agentic AI, AI, AI agents, AI coding, AI work, Anthropic, Biz & IT, Claude Code, Codex, context windows, large language models, machine learning, openai, Programming, software development, vibe coding / Paul Patrick / December 24, 2025

Agents of uncertain change

From compression tricks to multi-agent teamwork, here’s what makes them tick.

AI coding agents from OpenAI, Anthropic, and Google can now work on software projects for hours at a time, writing complete apps, running tests, and fixing bugs with human supervision. But these tools are not magic and can complicate rather than simplify a software project. Understanding how they work under the hood can help developers know when (and if) to use them, while avoiding common pitfalls.

We’ll start with the basics: At the core of every AI coding agent is a technology called a large language model (LLM), which is a type of neural network trained on vast amounts of text data, including lots of programming code. It’s a pattern-matching machine that uses a prompt to “extract” compressed statistical representations of data it saw during training and provide a plausible continuation of that pattern as an output. In this extraction, an LLM can interpolate across domains and concepts, resulting in some useful logical inferences when done well and confabulation errors when done poorly.

These base models are then further refined through techniques like fine-tuning on curated examples and reinforcement learning from human feedback (RLHF), which shape the model to follow instructions, use tools, and produce more useful outputs.

A screenshot of the Claude Code command-line interface. Credit: Anthropic

Over the past few years, AI researchers have been probing LLMs’ deficiencies and finding ways to work around them. One recent innovation was the simulated reasoning model, which generates context (extending the prompt) in the form of reasoning-style text that can help an LLM home in on a more accurate output. Another innovation was an application called an “agent” that links several LLMs together to perform tasks simultaneously and evaluate outputs.

How coding agents are structured

In that sense, each AI coding agent is a program wrapper that works with multiple LLMs. There is typically a “supervising” LLM that interprets tasks (prompts) from the human user and then assigns those tasks to parallel LLMs that can use software tools to execute the instructions. The supervising agent can interrupt tasks below it and evaluate the subtask results to see how a project is going. Anthropic’s engineering documentation describes this pattern as “gather context, take action, verify work, repeat.”

If run locally through a command-line interface (CLI), users give the agents conditional permission to write files on the local machine (code or whatever is needed), run exploratory commands (say, “ls” to list files in a directory), fetch websites (usually using “curl”), download software, or upload files to remote servers. There are lots of possibilities (and potential dangers) with this approach, so it needs to be used carefully.

In contrast, when a user starts a task in the web-based agent like the web versions of Codex and Claude Code, the system provisions a sandboxed cloud container preloaded with the user’s code repository, where Codex can read and edit files, run commands (including test harnesses and linters), and execute code in isolation. Anthropic’s Claude Code uses operating system-level features to create filesystem and network boundaries within which the agent can work more freely.

The context problem

Every LLM has a short-term memory, so to speak, that limits the amount of data it can process before it “forgets” what it’s doing. This is called “context.” Every time you submit a response to the supervising agent, you are amending one gigantic prompt that includes the entire history of the conversation so far (and all the code generated, plus the simulated reasoning tokens the model uses to “think” more about a problem). The AI model then evaluates this prompt and produces an output. It’s a very computationally expensive process that increases quadratically with prompt size because LLMs process every token (chunk of data) against every other token in the prompt.

Anthropic’s engineering team describes context as a finite resource with diminishing returns. Studies have revealed what researchers call “context rot”: As the number of tokens in the context window increases, the model’s ability to accurately recall information decreases. Every new token depletes what the documentation calls an “attention budget.”

This context limit naturally limits the size of a codebase a LLM can process at one time, and if you feed the AI model lots of huge code files (which have to be re-evaluated by the LLM every time you send another response), it can burn up token or usage limits pretty quickly.

Tricks of the trade

To get around these limits, the creators of coding agents use several tricks. For example, AI models are fine-tuned to write code to outsource activities to other software tools. For example, they might write Python scripts to extract data from images or files rather than feeding the whole file through an LLM, which saves tokens and avoids inaccurate results.

Anthropic’s documentation notes that Claude Code also uses this approach to perform complex data analysis over large databases, writing targeted queries and using Bash commands like “head” and “tail” to analyze large volumes of data without ever loading the full data objects into context.

(In a way, these AI agents are guided but semi-autonomous tool-using programs that are a major extension of a concept we first saw in early 2023.)

Another major breakthrough in agents came from dynamic context management. Agents can do this in a few ways that are not fully disclosed in proprietary coding models, but we do know the most important technique they use: context compression.

The command line version of OpenAI codex running in a macOS terminal window. — The command-line version of OpenAI Codex running in a macOS terminal window. Credit: Benj Edwards

When a coding LLM nears its context limit, this technique compresses the context history by summarizing it, losing details in the process but shortening the history to key details. Anthropic’s documentation describes this “compaction” as distilling context contents in a high-fidelity manner, preserving key details like architectural decisions and unresolved bugs while discarding redundant tool outputs.

This means the AI coding agents periodically “forget” a large portion of what they are doing every time this compression happens, but unlike older LLM-based systems, they aren’t completely clueless about what has transpired and can rapidly re-orient themselves by reading existing code, written notes left in files, change logs, and so on.

Anthropic’s documentation recommends using CLAUDE.md files to document common bash commands, core files, utility functions, code style guidelines, and testing instructions. AGENTS.md, now a multi-company standard, is another useful way of guiding agent actions in between context refreshes. These files act as external notes that let agents track progress across complex tasks while maintaining critical context that would otherwise be lost.

For tasks requiring extended work, both companies employ multi-agent architectures. According to Anthropic’s research documentation, its system uses an “orchestrator-worker pattern” in which a lead agent coordinates the process while delegating to specialized subagents that operate in parallel. When a user submits a query, the lead agent analyzes it, develops a strategy, and spawns subagents to explore different aspects simultaneously. The subagents act as intelligent filters, returning only relevant information rather than their full context to the lead agent.

The multi-agent approach burns through tokens rapidly. Anthropic’s documentation notes that agents typically use about four times more tokens than chatbot interactions, and multi-agent systems use about 15 times more tokens than chats. For economic viability, these systems require tasks where the value is high enough to justify the increased cost.

Best practices for humans

While using these agents is contentious in some programming circles, if you use one to code a project, knowing good software development practices helps to head off future problems. For example, it’s good to know about version control, making incremental backups, implementing one feature at a time, and testing it before moving on.

What people call “vibe coding”—creating AI-generated code without understanding what it’s doing—is clearly dangerous for production work. Shipping code you didn’t write yourself in a production environment is risky because it could introduce security issues or other bugs or begin gathering technical debt that could snowball over time.

Independent AI researcher Simon Willison recently argued that developers using coding agents still bear responsibility for proving their code works. “Almost anyone can prompt an LLM to generate a thousand-line patch and submit it for code review,” Willison wrote. “That’s no longer valuable. What’s valuable is contributing code that is proven to work.”

In fact, human planning is key. Claude Code’s best practices documentation recommends a specific workflow for complex problems: First, ask the agent to read relevant files and explicitly tell it not to write any code yet, then ask it to make a plan. Without these research and planning steps, the documentation warns, Claude’s outputs tend to jump straight to coding a solution.

Without planning, LLMs sometimes reach for quick solutions to satisfy a momentary objective that might break later if a project were expanded. So having some idea of what makes a good architecture for a modular program that can be expanded over time can help you guide the LLM to craft something more durable.

As mentioned above, these agents aren’t perfect, and some people prefer not to use them at all. A randomized controlled trial published by the nonprofit research organization METR in July 2025 found that experienced open-source developers actually took 19 percent longer to complete tasks when using AI tools, despite believing they were working faster. The study’s authors note several caveats: The developers were highly experienced with their codebases (averaging five years and 1,500 commits), the repositories were large and mature, and the models used (primarily Claude 3.5 and 3.7 Sonnet via Cursor) have since been superseded by more capable versions.

Whether newer models would produce different results remains an open question, but the study suggests that AI coding tools may not always provide universal speed-ups, particularly for developers who already know their codebases well.

Given these potential hazards, coding proof-of-concept demos and internal tools is probably the ideal use of coding agents right now. Since AI models have no actual agency (despite being called agents) and are not people who can be held accountable for mistakes, human oversight is key.

How AI coding agents work—and what to remember if you use them Read More »

OpenAI built an AI coding agent and uses it to improve the agent itself

agentic AI, AI, AI agents, AI assistants, AI coding, AI development tools, AI programming, AI work, Anthropic, Biz & IT, chatbots, Claude, code agents, Codex, Dario Amodei, generative ai, GitHub Copilot, Google, GPT-5, large language models, machine learning, Mistral, Open Source, openai, Programming, sam altman, software development, vibe coding / Kris Guyer / December 12, 2025

“The vast majority of Codex is built by Codex,” OpenAI told us about its new AI coding agent.

With the popularity of AI coding tools rising among software developers, their adoption has begun to touch every aspect of the process, including the improvement of AI coding tools themselves.

In interviews with Ars Technica this week, OpenAI employees revealed the extent to which the company now relies on its own AI coding agent, Codex, to build and improve the development tool. “I think the vast majority of Codex is built by Codex, so it’s almost entirely just being used to improve itself,” said Alexander Embiricos, product lead for Codex at OpenAI, in a conversation on Tuesday.

Codex, which OpenAI launched in its modern incarnation as a research preview in May 2025, operates as a cloud-based software engineering agent that can handle tasks like writing features, fixing bugs, and proposing pull requests. The tool runs in sandboxed environments linked to a user’s code repository and can execute multiple tasks in parallel. OpenAI offers Codex through ChatGPT’s web interface, a command-line interface (CLI), and IDE extensions for VS Code, Cursor, and Windsurf.

The “Codex” name itself dates back to a 2021 OpenAI model based on GPT-3 that powered GitHub Copilot’s tab completion feature. Embiricos said the name is rumored among staff to be short for “code execution.” OpenAI wanted to connect the new agent to that earlier moment, which was crafted in part by some who have left the company.

“For many people, that model powering GitHub Copilot was the first ‘wow’ moment for AI,” Embiricos said. “It showed people the potential of what it can mean when AI is able to understand your context and what you’re trying to do and accelerate you in doing that.”

A place to enter a prompt, set parameters, and click — The interface for OpenAI’s Codex in ChatGPT. Credit: OpenAI

It’s no secret that the current command-line version of Codex bears some resemblance to Claude Code, Anthropic’s agentic coding tool that launched in February 2025. When asked whether Claude Code influenced Codex’s design, Embiricos parried the question but acknowledged the competitive dynamic. “It’s a fun market to work in because there’s lots of great ideas being thrown around,” he said. He noted that OpenAI had been building web-based Codex features internally before shipping the CLI version, which arrived after Anthropic’s tool.

OpenAI’s customers apparently love the command line version, though. Embiricos said Codex usage among external developers jumped 20 times after OpenAI shipped the interactive CLI extension alongside GPT-5 in August 2025. On September 15, OpenAI released GPT-5 Codex, a specialized version of GPT-5 optimized for agentic coding, which further accelerated adoption.

It hasn’t just been the outside world that has embraced the tool. Embiricos said the vast majority of OpenAI’s engineers now use Codex regularly. The company uses the same open-source version of the CLI that external developers can freely download, suggest additions to, and modify themselves. “I really love this about our team,” Embiricos said. “The version of Codex that we use is literally the open source repo. We don’t have a different repo that features go in.”

The recursive nature of Codex development extends beyond simple code generation. Embiricos described scenarios where Codex monitors its own training runs and processes user feedback to “decide” what to build next. “We have places where we’ll ask Codex to look at the feedback and then decide what to do,” he said. “Codex is writing a lot of the research harness for its own training runs, and we’re experimenting with having Codex monitoring its own training runs.” OpenAI employees can also submit a ticket to Codex through project management tools like Linear, assigning it tasks the same way they would assign work to a human colleague.

This kind of recursive loop, of using tools to build better tools, has deep roots in computing history. Engineers designed the first integrated circuits by hand on vellum and paper in the 1960s, then fabricated physical chips from those drawings. Those chips powered the computers that ran the first electronic design automation (EDA) software, which in turn enabled engineers to design circuits far too complex for any human to draft manually. Modern processors contain billions of transistors arranged in patterns that exist only because software made them possible. OpenAI’s use of Codex to build Codex seems to follow the same pattern: each generation of the tool creates capabilities that feed into the next.

But describing what Codex actually does presents something of a linguistic challenge. At Ars Technica, we try to reduce anthropomorphism when discussing AI models as much as possible while also describing what these systems do using analogies that make sense to general readers. People can talk to Codex like a human, so it feels natural to use human terms to describe interacting with it, even though it is not a person and simulates human personality through statistical modeling.

The system runs many processes autonomously, addresses feedback, spins off and manages child processes, and produces code that ships in real products. OpenAI employees call it a “teammate” and assign it tasks through the same tools they use for human colleagues. Whether the tasks Codex handles constitute “decisions” or sophisticated conditional logic smuggled through a neural network depends on definitions that computer scientists and philosophers continue to debate. What we can say is that a semi-autonomous feedback loop exists: Codex produces code under human direction, that code becomes part of Codex, and the next version of Codex produces different code as a result.

Building faster with “AI teammates”

According to our interviews, the most dramatic example of Codex’s internal impact came from OpenAI’s development of the Sora Android app. According to Embiricos, the development tool allowed the company to create the app in record time.

“The Sora Android app was shipped by four engineers from scratch,” Embiricos told Ars. “It took 18 days to build, and then we shipped it to the app store in 28 days total,” he said. The engineers already had the iOS app and server-side components to work from, so they focused on building the Android client. They used Codex to help plan the architecture, generate sub-plans for different components, and implement those components.

Despite OpenAI’s claims of success with Codex in house, it’s worth noting that independent research has shown mixed results for AI coding productivity. A METR study published in July found that experienced open source developers were actually 19 percent slower when using AI tools on complex, mature codebases—though the researchers noted AI may perform better on simpler projects.

Ed Bayes, a designer on the Codex team, described how the tool has changed his own workflow. Bayes said Codex now integrates with project management tools like Linear and communication platforms like Slack, allowing team members to assign coding tasks directly to the AI agent. “You can add Codex, and you can basically assign issues to Codex now,” Bayes told Ars. “Codex is literally a teammate in your workspace.”

This integration means that when someone posts feedback in a Slack channel, they can tag Codex and ask it to fix the issue. The agent will create a pull request, and team members can review and iterate on the changes through the same thread. “It’s basically approximating this kind of coworker and showing up wherever you work,” Bayes said.

For Bayes, who works on the visual design and interaction patterns for Codex’s interfaces, the tool has enabled him to contribute code directly rather than handing off specifications to engineers. “It kind of gives you more leverage. It enables you to work across the stack and basically be able to do more things,” he said. He noted that designers at OpenAI now prototype features by building them directly, using Codex to handle the implementation details.

OpenAI’s approach treats Codex as what Bayes called “a junior developer” that the company hopes will graduate into a senior developer over time. “If you were onboarding a junior developer, how would you onboard them? You give them a Slack account, you give them a Linear account,” Bayes said. “It’s not just this tool that you go to in the terminal, but it’s something that comes to you as well and sits within your team.”

Given this teammate approach, will there be anything left for humans to do? When asked, Embiricos drew a distinction between “vibe coding,” where developers accept AI-generated code without close review, and what AI researcher Simon Willison calls “vibe engineering,” where humans stay in the loop. “We see a lot more vibe engineering in our code base,” he said. “You ask Codex to work on that, maybe you even ask for a plan first. Go back and forth, iterate on the plan, and then you’re in the loop with the model and carefully reviewing its code.”

He added that vibe coding still has its place for prototypes and throwaway tools. “I think vibe coding is great,” he said. “Now you have discretion as a human about how much attention you wanna pay to the code.”

Looking ahead

Over the past year, “monolithic” large language models (LLMs) like GPT-4.5 have apparently become something of a dead end in terms of frontier benchmarking progress as AI companies pivot to simulated reasoning models and also agentic systems built from multiple AI models running in parallel. We asked Embiricos whether agents like Codex represent the best path forward for squeezing utility out of existing LLM technology.

He dismissed concerns that AI capabilities have plateaued. “I think we’re very far from plateauing,” he said. “If you look at the velocity on the research team here, we’ve been shipping models almost every week or every other week.” He pointed to recent improvements where GPT-5-Codex reportedly completes tasks 30 percent faster than its predecessor at the same intelligence level. During testing, the company has seen the model work independently for 24 hours on complex tasks.

OpenAI faces competition from multiple directions in the AI coding market. Anthropic’s Claude Code and Google’s Gemini CLI offer similar terminal-based agentic coding experiences. This week, Mistral AI released Devstral 2 alongside a CLI tool called Mistral Vibe. Meanwhile, startups like Cursor have built dedicated IDEs around AI coding, reportedly reaching $300 million in annualized revenue.

Given the well-known issues with confabulation in AI models when people attempt to use them as factual resources, could it be that coding has become the killer app for LLMs? We wondered if OpenAI has noticed that coding seems to be a clear business use case for today’s AI models with less hazard than, say, using AI language models for writing or as emotional companions.

“We have absolutely noticed that coding is both a place where agents are gonna get good really fast and there’s a lot of economic value,” Embiricos said. “We feel like it’s very mission-aligned to focus on Codex. We get to provide a lot of value to developers. Also, developers build things for other people, so we’re kind of intrinsically scaling through them.”

But will tools like Codex threaten software developer jobs? Bayes acknowledged concerns but said Codex has not reduced headcount at OpenAI, and “there’s always a human in the loop because the human can actually read the code.” Similarly, the two men don’t project a future where Codex runs by itself without some form of human oversight. They feel the tool is an amplifier of human potential rather than a replacement for it.

The practical implications of agents like Codex extend beyond OpenAI’s walls. Embiricos said the company’s long-term vision involves making coding agents useful to people who have no programming experience. “All humanity is not gonna open an IDE or even know what a terminal is,” he said. “We’re building a coding agent right now that’s just for software engineers, but we think of the shape of what we’re building as really something that will be useful to be a more general agent.”

This article was updated on December 12, 2025 at 6: 50 PM to mention the METR study.

OpenAI built an AI coding agent and uses it to improve the agent itself Read More »

Big Tech joins forces with Linux Foundation to standardize AI agents

AI, Anthropic, Artificial Intelligence, MCP, openai / Paul Patrick / December 9, 2025

Big Tech has spent the past year telling us we’re living in the era of AI agents, but most of what we’ve been promised is still theoretical. As companies race to turn fantasy into reality, they’ve developed a collection of tools to guide the development of generative AI. A cadre of major players in the AI race, including Anthropic, Block, and OpenAI, has come together to promote interoperability with the newly formed Agentic AI Foundation (AAIF). This move elevates a handful of popular technologies and could make them a de facto standard for AI development going forward.

The development path for agentic AI models is cloudy to say the least, but companies have invested so heavily in creating these systems that some tools have percolated to the surface. The AAIF, which is part of the nonprofit Linux Foundation, has been launched to govern the development of three key AI technologies: Model Context Protocol (MCP), goose, and AGENTS.md.

MCP is probably the most well-known of the trio, having been open-sourced by Anthropic a year ago. The goal of MCP is to link AI agents to data sources in a standardized way—Anthropic (and now the AAIF) is fond of calling MCP a “USB-C port for AI.” Rather than creating custom integrations for every different database or cloud storage platform, MCP allows developers to quickly and easily connect to any MCP-compliant server.

Since its release, MCP has been widely used across the AI industry. Google announced at I/O 2025 that it was adding support for MCP in its dev tools, and many of its products have since added MCP servers to make data more accessible to agents. OpenAI also adopted MCP just a few months after it was released.

Expanding use of MCP might help users customize their AI experience. For instance, the new Pebble Index 01 ring uses a local LLM that can act on your voice notes, and it supports MCP for user customization.

Local AI models have to make some sacrifices compared to bigger cloud-based models, but MCP can fill in the functionality gaps. “A lot of tasks on productivity and content are fully doable on the edge,” Qualcomm head of AI products, Vinesh Sukumar, tells Ars. “With MCP, you have a handshake with multiple cloud service providers for any kind of complex task to be completed.”

Big Tech joins forces with Linux Foundation to standardize AI agents Read More »

Anthropic introduces cheaper, more powerful, more efficienct Opus 4.5 model

AI, Anthropic, Claude, Claude API, Claude Code, large language model, llm, Opus 4.5 / Paul Patrick / November 24, 2025

Anthropic today released Opus 4.5, its flagship frontier model, and it brings improvements in coding performance, as well as some user experience improvements that make it more generally competitive with OpenAI’s latest frontier models.

Perhaps the most prominent change for most users is that in the consumer app experiences (web, mobile, and desktop), Claude will be less prone to abruptly hard-stopping conversations because they have run too long. The improvement to memory within a single conversation applies not just to Opus 4.5, but to any current Claude models in the apps.

Users who experienced abrupt endings (despite having room left in their session and weekly usage budgets) were hitting a hard context window (200,000 tokens). Whereas some large language model implementations simply start trimming earlier messages from the context when a conversation runs past the maximum in the window, Claude simply ended the conversation rather than allow the user to experience an increasingly incoherent conversation where the model would start forgetting things based on how old they are.

Now, Claude will instead go through a behind-the-scenes process of summarizing the key points from the earlier parts of the conversation, attempting to discard what it deems extraneous while keeping what’s important.

Developers who call Anthropic’s API can leverage the same principles through context management and context compaction.

Opus 4.5 performance

Opus 4.5 is the first model to surpass an accuracy score of 80 percent—specifically, 80.9 percent in the SWE-Bench Verified benchmark, narrowly beating OpenAI’s recently released GPT-5.1-Codex-Max (77.9 percent) and Google’s Gemini 3 Pro (76.2 percent). The model performs particularly well in agentic coding and agentic tool use benchmarks, but still lags behind GPT-5.1 in visual reasoning (MMMU).

Anthropic introduces cheaper, more powerful, more efficienct Opus 4.5 model Read More »

Tech giants pour billions into Anthropic as circular AI investments roll on

AI, AI bubble, AI chips, AI infrastructure, Anthropic, azure, Biz & IT, Cloud Computing, datacenters, enterprise AI, machine learning, microsoft, NVIDIA, openai, sam altman, Satya Nadella / Mike M. / November 18, 2025

On Tuesday, Microsoft and Nvidia announced plans to invest in Anthropic under a new partnership that includes a $30 billion commitment by the Claude maker to use Microsoft’s cloud services. Nvidia will commit up to $10 billion to Anthropic and Microsoft up to $5 billion, with both companies investing in Anthropic’s next funding round.

The deal brings together two companies that have backed OpenAI and connects them more closely to one of the ChatGPT maker’s main competitors. Microsoft CEO Satya Nadella said in a video that OpenAI “remains a critical partner,” while adding that the companies will increasingly be customers of each other.

“We will use Anthropic models, they will use our infrastructure, and we’ll go to market together,” Nadella said.

Anthropic, Microsoft, and NVIDIA announce partnerships.

The move follows OpenAI’s recent restructuring that gave the company greater distance from its non-profit origins. OpenAI has since announced a $38 billion deal to buy cloud services from Amazon.com as the company becomes less dependent on Microsoft. OpenAI CEO Sam Altman has said the company plans to spend $1.4 trillion to develop 30 gigawatts of computing resources.

Tech giants pour billions into Anthropic as circular AI investments roll on Read More »

Researchers question Anthropic claim that AI-assisted attack was 90% autonomous

AI, Anthropic, Biz & IT, Claude, Cyberattacks, espionage, Security / Mike M. / November 15, 2025

Claude frequently overstated findings and occasionally fabricated data during autonomous operations, claiming to have obtained credentials that didn’t work or identifying critical discoveries that proved to be publicly available information. This AI hallucination in offensive security contexts presented challenges for the actor’s operational effectiveness, requiring careful validation of all claimed results. This remains an obstacle to fully autonomous cyberattacks.

How (Anthropic says) the attack unfolded

Anthropic said GTG-1002 developed an autonomous attack framework that used Claude as an orchestration mechanism that largely eliminated the need for human involvement. This orchestration system broke complex multi-stage attacks into smaller technical tasks such as vulnerability scanning, credential validation, data extraction, and lateral movement.

“The architecture incorporated Claude’s technical capabilities as an execution engine within a larger automated system, where the AI performed specific technical actions based on the human operators’ instructions while the orchestration logic maintained attack state, managed phase transitions, and aggregated results across multiple sessions,” Anthropic said. “This approach allowed the threat actor to achieve operational scale typically associated with nation-state campaigns while maintaining minimal direct involvement, as the framework autonomously progressed through reconnaissance, initial access, persistence, and data exfiltration phases by sequencing Claude’s responses and adapting subsequent requests based on discovered information.”

The attacks followed a five-phase structure that increased AI autonomy through each one.

The life cycle of the cyberattack, showing the move from human-led targeting to largely AI-driven attacks using various tools, often via the Model Context Protocol (MCP). At various points during the attack, the AI returns to its human operator for review and further direction. Credit: Anthropic

The attackers were able to bypass Claude guardrails in part by breaking tasks into small steps that, in isolation, the AI tool didn’t interpret as malicious. In other cases, the attackers couched their inquiries in the context of security professionals trying to use Claude to improve defenses.

As noted last week, AI-developed malware has a long way to go before it poses a real-world threat. There’s no reason to doubt that AI-assisted cyberattacks may one day produce more potent attacks. But the data so far indicates that threat actors—like most others using AI—are seeing mixed results that aren’t nearly as impressive as those in the AI industry claim.

Researchers question Anthropic claim that AI-assisted attack was 90% autonomous Read More »