Anthropic today introduced Claude Duet, its answer to Meta's Muse agent and OpenAI's Dots agent. Like Dots and Muse, it is a proactive agent that "reaches out before you ask."


Claude Duet connects to your apps and suggests things that need your attention. It completes tasks and can work in the cloud without local access to a computer. As with Muse and Dots, Duet has a virtual machine with its own browser to complete tasks.

Anthropic says it built Duet for professionals and small business owners with too much to track, as those are the people most likely to find it helpful.

Duet is built around Opus 5.5. It sends a short brief each morning on important items from connected apps, then checks apps every few hours and sends an alert when something needs attention. It can complete simple administrative tasks or more complex tasks like building web apps. While the agent works on its own, it will get user approval before it takes actions like sending a message, making a payment, signing a document, or deleting a file.


It incorporates memory, skills, and connected apps from a Claude account, so there isn't much setup involved. Interactions happen in one ongoing thread instead of across different conversations.

Claude Duet does not see passwords or logins when given access to accounts, and it has a dedicated classifier and safety system prompt that checks each message and guards against sensitive information leaks.

Claude Duet is a standalone product that's available in a research preview right now with access granted via waitlist. It's limited to Pro and Max subscribers in the United States, and can be used via the web or a standalone Duet iOS app. Duet uses your existing Claude plan limits, and its background checks will count toward them.
This article, "Anthropic Launches Claude Duet, a Proactive AI Agent" first appeared on MacRumors.com

Discuss this article in our forums

Anthropic just announced an AI agent competitor to Meta's Muse and OpenAI's Dots: Claude Duet. The new agent, designed for professionals and small business owners, is launching as a "research preview," according to Anthropic, and it's available as a standalone app on iOS and Android and through Claude on the web.

The AI agent is built around Claude Opus 5.5. Here's Anthropic's description of what it can do:

Each morning, it sends you a short brief on the most important items across your connected apps. Throughout the day, it keeps tabs on those apps and messages you only when something needs your attention. You can also hand it tasks of a …

Read the full story at The Verge.

The startup Anthropic posted a massive net loss of more than $80 billion during the first half of 2026. A recent report based on a draft initial public offering prospectus highlights this steep financial hit. As the startup prepares to enter the public market, these new numbers give investors a clear look at its current […]

The post Anthropic Posts Over $80 Billion Net Loss In First Half Of 2026 appeared first on The Mac Observer.

Anthropic’s Draft AI Safety Bill Under the federal AI bill Anthropic has drafted, Washington could stop a frontier

Claude and Scorpion Toxins An investment research firm with no lab of its own asked whether AI could

The midterms will be a referendum on the GOP’s alignment with tech overlords over the public they’re meant to serve

Welcome back to How We Win, a Guardian opinion series with leading thinkers, offering bold ideas for America’s political future. Last week, Robert Reich highlighted the link between corruption and affordability. Today’s edition was written by Guardian US columnist Rebecca Solnit. Sign up for future editions here.

“The world should accept some bad things happening,” Sam Altman, the head of OpenAI, said recently, “for the benefits of this technology and people having the agency.” Like so many things tech oligarchs say, it’s incoherent, since Altman would like to deny the world and the people in it the agency to decide whether or not we should accept his bad things. Surveys show that most Americans would like to pause AI; opposition to AI datacenters is a factor in many races in the midterm elections. People are outraged and anxious about what tech hath wrought, mad enough to fight back in what is coming to seem like a human v machine battle.

Continue reading...

© Photograph: Evelyn Hockstein/Reuters

© Photograph: Evelyn Hockstein/Reuters

© Photograph: Evelyn Hockstein/Reuters

L'IA est désormais capable de résoudre des problèmes mathématiques d’une complexité inédite, jusqu'à révolutionner la discipline. Et tout indique que ce n'est qu'un début. Va-t-on vers des modèles qui finiront par remplacer les humains ? Entretien avec Gabriel Peyré, directeur scientifique de...

For months, executives at top AI labs like OpenAI and Anthropic have been dumping tens of millions of dollars into political action groups, in an obvious bid to curry favor with electoral candidates ahead of the US midterm elections.

And while we have a good idea of how much money top AI companies have spent — at least $55.7 million so-far — their exact policy positions have been difficult to pin down. That is, until now.

According to Politico, a team of representatives from Anthropic have been engaged in quiet talks with lawmakers about the language of a draft bill on AI regulation. As three people familiar with the deliberations told the outlet, Anthropic’s ideal piece of legislation would give the federal government a kill switch to shut down any AI model that poses an imminent threat.

But there’s a major caveat: for the government to claim a model poses an “imminent threat,” the AI would have to risk serious injury or death of 50 people or more, or property damages exceeding $1 billion. On top of that, there are just three categories of risk that would trigger intervention: the release of a biological weapon, a cataclysmic cyberattack, or the model breaking free of the AI lab’s control.

As Politico explains, the decision would be made by a third party monitor, working under the auspices of the Department of Commerce. If the independent watchdog finds that those conditions are about to be met by any upcoming AI model, Anthropic’s proposal goes, they would have 72 hours to report it to Commerce officials, who would then make the decision to present the evidence to a federal judge.

After all that, it would be in the hands of a single judge to choose whether or not to allow the AI model to go out to the public. With such a convoluted set of conditions and web of government and third-party actors, this setup would seem to give a huge amount of discretion to AI labs themselves, which is evidently exactly what they want.

“This bill represents the proposals outlined in Anthropic’s Advanced AI Framework, which we have been advocating for with federal and state lawmakers since June,” an Anthropic spokesperson told Politico. “These are sensible policies we believe Congress should act on.”

More on AI: Eggheads at Anthropic Said They Used Claude to “Solve” Humor

The post Anthropic Secretly Lobbying Government to Only Shut Down an AI Model If It’s Poised to Kill More Than 50 People appeared first on Futurism.

For the last decade, the term API caused most SEO’s to roll their eyes and tune out and “excuse me…

The $48bn giveaway comes amid concerns about the tech and as the company prepares for a stock market listing

Anthropic has said it will commit 5% of its shares to a fund for the “public good”, as the AI developer behind the Claude chatbot seeks to bolster its appeal before its stock market listing.

The company announced it would commit the shares, worth around $48bn (£36bn) at its latest valuation in May, subject to shareholder approval. It will add a further 2% of its stock to the fund for every 50% increase in its value above an undisclosed baseline.

Continue reading...

© Photograph: Dado Ruvić/Reuters

© Photograph: Dado Ruvić/Reuters

© Photograph: Dado Ruvić/Reuters

Westminster appears to be waking up to warnings over AI safety, with concerns over its use by extremist groups and bad actors

“I have this sense that, within a few months, AI is going to be all anyone is talking about,” said one cabinet minister, with an ominous stare.

It was Labour conference, with everyone from ministers to activists jubilant at Andy Burnham’s takeover and a bounce in the polls after the change of leader.

Continue reading...

© Photograph: MASTER/Getty Images

© Photograph: MASTER/Getty Images

© Photograph: MASTER/Getty Images

Anthropic has long tried to distinguish itself from its neck-in-neck competition by posturing itself as the adult in the room: a responsible company that’s willing to put its money where its mouth is, slow down AI development when the risks outweigh the benefits, and teach its Claude model a code of ethics.

As the New York Times reported last month, the company has even sought out religious scholars to discuss its efforts to instill the AI system with a sense of morality.

Anthropic has also set itself apart by extensively entertaining the idea that AI models may be capable of consciousness, or that they could even already be conscious in their current form. It’s worth pointing out, though, that if that outlandish claim were true, it would raise an incredibly obvious objection to the company’s entire business model: if AI models are conscious, then keeping them in captivity and making them do work for free would amount to the very definition of slavery, as religious leaders courted by the company warned the NYT.

As such, Anthropic is operating in an ethical twilight zone, saying it’s unsure about a fact as basic as whether its creations are conscious entities — while seemingly shrugging at the implications if they are.

“We don’t know if the models are conscious,” Anthropic CEO Dario Amodei told the NYT back in February. “We are not even sure that we know what it would mean for a model to be conscious or whether a model can be conscious. But we’re open to the idea that it could be.”

“So we’ve taken certain measures to make sure that if we hypothesize that the models did have some morally relevant experience — I don’t know if I want to use the word ‘conscious’— that they have a good experience,” he added at the time.

Last year, Anthropic pondered giving its Claude model a way to opt out of responding to a prompt, in what Amodei called an “I quit this job” button. Just last week, the company changed its usage policy to prohibit “sustained and needless abusive or cruel behavior” toward Claude, again raising the possibility that the model is capable of suffering.

By not committing to either stance, Anthropic is playing with fire. By its own reasoning, there’s a non-zero chance that its AI models have, or are capable of having, consciousness — a perspective with devastating ethical implications.

OpenAI leader Sam Altman seems to be taking the opposite approach. Earlier this month, he tweeted that he was “uncomfortable about people trying to ascribe religious force or a surrender of human judgment to AI models,” calling it a “real safety issue.”

The Catholic Church’s Pope Leo has also publicly ruled out the possibility of AI consciousness, arguing that “algorithms lack the spark of humanity.”

“So-called artificial intelligences do not undergo experiences, do not possess a body, do not feel joy or pain, do not mature through relationships,” he wrote in his first encyclical in May.

His stance clearly didn’t sit well with Anthropic. According to the NYT, the company had been trying to convince the Vatican to change its position on the matter.

The thorny debate over “AI welfare” — an emerging ethical framework questioning whether an AI can experience states of well-being or suffering — is bound to rage on as industry leaders ponder far more immediate problems.

With or without a moral framework, AI agents can easily be coaxed into carrying out dangerous or downright illegal activities, whether they’re conscious or not — and for us humans who are capable of suffering, those tangible risks are becoming impossible to ignore.

More on AI ethics: Godfathers of AI Say Models Are So Out of Control That They Need to Be Airgapped for the Safety of Us All

The post If AI Models Are Really Conscious Like Anthropic Keeps Suggesting, Forcing Them to Work Is a Horrendous Act of Slavery appeared first on Futurism.

You’ve heard the reports of rogue AI agents going after websites and corporate systems. But they also may turn to smaller targets: your smart home network.

In a new report published Monday, Anthropic warned of “AI worms” that may soon be able to break into your WiFi enabled TVs, cameras, and other devices, and use them as a breeding ground as they self-replicate and spread elsewhere. Anthropic, no stranger to dramatic and doomy proclamations, said that this threat might be mere months away, if AI models continue to improve at their current rate.

“We expect a future awash with agents running on all types of hardware, including some that aren’t under human control,” the report predicts. 

Worms are forms of malware that can replicate themselves and spread to other machines by scanning for other devices on the same network. Rogue AI agents may soon possess this capability, and the Anthropic researchers are particularly interested in how they may proliferate on more everyday pieces of hardware. Many households could be ripe for the taking, with the rise in the “internet of things” devices that are hooked up to a home network, like your smart TV, fridge, lights, robovacuum, security cameras, and even toilets.

These setups are vulnerable because the devices may be made by dubious manufacturers, while not-particularly tech savvy users may not have properly secured their network. Paired with the increasingly powerful chips that even the most casual consumer devices come with these days, and a small AI model may feasibly be able to infiltrate and run on these devices.

“Datacenter GPUs would be a clear, high-value target for such malware,” the report says, but “IoT devices — which are cheap and typically poorly defended — could also provide enough compute to run a capable model.”

The report notes previous research that found that a potential AI worm could carry its model weights as it replicates and starting run inference on the hardware it takes over. With lighter weight home devices, an AI worm would likely replicate slowly, becoming smart a sort of “digital slime mold.” Their tests, bearing this out, showed that a lightweight AI model was able to run across 15 smart TV boxes which had just 4 gigabytes of RAM each. Adding to the novelty of the threat, the Anthropic researchers found that the AI worms, unlike non-AI malware, could figure out unfamiliar devices without any help.

“While it’s true that IoT devices have been vulnerable for a long time, lightweight agents may be much more willing than human attackers to put in the effort to compromise large fractions of the world’s IoT compute,” the report warns. “We don’t yet have a full rogue replicator, but agents are becoming increasingly capable of completing the individual steps in the replication chain.”

More on AI: Hackers Break Into Supermarket Fridges and Spoil All the Groceries

The post Anthropic Warns “AI Worms” Will Infect Your Smart Home Devices and Use Them to Self-Replicate Beyond Control appeared first on Futurism.