OpenClaw's continued buzz has kicked off a new race, with Anthropic, Nvidia, Perplexity and others all fast-tracking autonomous bots that can make OpenClaw's magic more palatable to businesses.

Why it matters: Companies are giving AI agents the ability to send emails, move files and change live systems — increasing both productivity and risk.

• "Autonomy only works if it's clear who can act, what's allowed, and how those decisions are tracked," Nick Durkin, CTO of software delivery platform Harness, told Axios.
• "Most companies are still figuring that part out."

Catch up quick: Anthropic in January released Claude Cowork, an AI agent that works with your files and tools directly for work tasks.

• OpenClaw launched before Cowork, but Cowork's big splash — especially among insiders — drew more users to OpenClaw's open source framework that set AI agents free, with minimal guardrails.

Driving the news: Now the excitement around OpenClaw has prompted companies to announce complementary products or rival claw-like systems.

• "Every single company" needs an "OpenClaw strategy," Nvidia CEO Jensen Huang said at last week's GTC conference in San Jose.

State of play: Nvidia last week debuted NemoClaw, a set of services it says can make OpenClaw more reliable and secure.

• Anthropic last week released Dispatch, a feature that allows Claude Cowork tasks to be launched from anywhere via a phone or other device while it runs on your local machine.
• "This is OpenClaw for grown-ups," Authority Hacker co-founder Gael Breton wrote on X. "It can do 90% [of] what OpenClaw does in a 90% more secure way."
• Perplexity used its first developer conference to pitch itself as a more secure alternative to OpenClaw. The company announced a business-centered version of its Perplexity Computer agent system and previewed Personal Computer, which runs on a Mac and has access to local files.
• Snowflake, the cloud-based data platform, released a similar autonomous platform for office tasks called Project SnowWork.

Zoom in: Agents designed for the enterprise can still go rogue.

• This week Meta confirmed to Axios that one of its in-house agents (similar to OpenClaw) posted advice in an internal forum without approval from a Meta employee.
• Another employee then acted on that advice — according to The Information — triggering a security incident that granted employees access to sensitive company and customer information.
• Meta says there is no evidence that any employees accessed that data.

What they're saying: "The engineering teams using AI most aggressively are experiencing more deployment failures and security incidents, not fewer," Durkin said.

• "More capability without more governance doesn't reduce risk. It just makes the problems harder to find."
• "At the end of the day, companies are going to be responsible for the actions of their agents, just like they're responsible for the actions of their employees," said Brooke Johnson, chief legal officer at Ivanti.
• The best advice is to treat AI like you would a human employee, but one that only understands rules, not morals, she said.

Threat level: Companies need to be very specific and intentional with both the tasks they give to agents and what systems they allow the agent to access, says James Everingham, CEO of Guild.ai, a startup that helps companies manage their agents.

• Agents will use all the access they have to achieve a goal, "whether it's right or wrong," Everingham tells Axios.

The bottom line: Companies shouldn't avoid using AI agents, but they should limit the tools and data agents have access to, experts told Axios.

OpenAI is in advanced talks to buy electricity from Sam Altman-backed fusion startup Helion Energy, according to a person familiar with the situation.

Why it matters: The move shows OpenAI is serious about tapping fusion energy to help meet massive power demands.

Driving the news: OpenAI could secure a guaranteed portion of Helion's production, initially 12.5%, per the source.

• The talks center on OpenAI receiving the equivalent of 5 gigawatts by 2030, scaling to 50 gigawatts by 2035.

Between the lines: OpenAI CEO Altman, who is an investor in Helion, has stepped down as Helion's board chair and is no longer involved in the company's board, the source told Axios.

• Altman has also recused himself from the deal discussions, according to the same source.
• Altman holds a sizable stake in Helion, though the size has not been disclosed. He led the company's $500 million Series E round in 2021. Helion closed a further $425 million funding round in January 2025.
• Helion is also backed by SoftBank, Peter Thiel's Mithril Capital and Facebook co-founder Dustin Moskovitz.

How it works: Fusion energy — which creates energy using the same process as the Sun — could be a key step in giving AI companies an abundant clean energy source to match its needs.

• Helion believes it is on the cusp of "scientific breakeven" — a key step to reaching the point where the company's fusion process can generate more energy than it consumes, the source said.
• No private company has yet achieved this milestone.

Yes, but: The talks with OpenAI involve the basic framework of an agreement, with many conditions yet to be fulfilled, including the selection of a site for Helion to produce the energy.

• An OpenAI representative declined to comment.

The big picture: Google has several agreements with Helion rival Commonwealth Fusion Systems, including a deal to buy 200 megawatts worth of power.

• Helion also inked a power purchase deal with Microsoft in 2023, calling for it to deliver 50 megawatts to the software giant.

What we're watching: Whether Helion can hit scientific breakeven on a timeline that makes those 2030 and 2035 power commitments a reality.

Nvidia and startup Emerald AI said today they're working with major U.S. energy companies to develop a new class of data centers designed to flex their power use and connect to the grid faster.

Why it matters: The effort reflects a growing push to turn AI data centers from massive power consumers into more dynamic grid participants, as electricity demand from AI surges.

How it works: These facilities could ramp power use up or down during periods of grid stress, similar to demand response programs, rather than running at constant full capacity.

Keep reading.

• Wired's Steven Levy has an inside look at Palantir's recent developer conference.
• OpenAI aims to grow to about 8,000 employees by year's end, up from around 4,500 as it looks to capture more of the dollars that businesses are spending on AI. (Financial Times)
• Workers are "tokenmaxxing" — competing against their colleagues to see who can consume the most AI tokens. (NYT)

A possum was found hanging out with stuffed toys in an airport gift shop in Tasmania.