Sam Altman’s Orb Was Built for the Bot Era. So Why Isn’t It Everywhere?
Welcome back to In the Loop, TIME’s twice-weekly newsletter about AI. We’re publishing these editions both as stories on Time.com and as emails. If you’re reading this in your browser, why not subscribe to have the next one delivered straight to your inbox?
What to Know: Is Sam Altman’s Orb missing its moment?
[time-brightcove not-tgx=”true”]When Moltbook, a social network for AI agents, went viral earlier this month, it should have been a vindication moment for Tools for Humanity — the startup co-founded by Sam Altman, whose eyeball-scanning “Orb” was designed to solve exactly this kind of problem. Instead, it may have exposed the product’s limitations.
For the uninitiated, the Orb is a spherical piece of hardware whose ostensible purpose is to protect the internet from the influx of AI. After scanning users’ irises, the Orb furnishes them with a code that can serve as their “proof of humanity.” Oh, and Tools for Humanity is also a crypto purveyor that flirted with, and then ditched, the idea of providing a universal basic income.
Got all that? I profiled Tools for Humanity last year, and I still struggle to succinctly explain what this project is, and what it’s for. But that process has become a little easier in the last few weeks, according to Trevor Traina, its chief business officer.
The Moltbook effect — Traina tells me that Moltbook’s viral moment led to a surge in inbound calls from companies seeking Tools for Humanity’s services. “The people who were calling us wacky are now just calling us,” Traina says. He says that businesses are increasingly coming to Tools for Humanity looking for a system that can distinguish between bots and humans online — exactly what the Orb was designed for. “When you see a network where autonomous bots are talking to each other, that’s the aha moment,” Traina says.
Wait a sec — I asked a company spokesperson to share data backing up the claim that business is booming, but they did not send any over by the time of publication. It’s also worth pausing to question what Moltbook really signifies. As the hype around the AI-only social network has died down, questions have surfaced about whether the site is really a signal that agents are becoming more autonomous, or rather, a form of “theater” that reveals the enduring clunkiness of today’s AI agents.
Awkward timing — You’ve got to hand it to Tools for Humanity: they correctly predicted, seven years ago, that AI agents would begin to overrun the internet. But despite that foresight, the company is now at risk of missing its moment. “We acknowledge we have a problem,” Traina says. “We’ve been Orb-constrained.” Nearly a year ago, the company announced that Orbs would be launching in the U.S., with 7,500 intended to be available across the country by this May. But at the very moment that AI agents are taking off, the Orb’s cultural impact remains negligible. (Convincing people to give up their eyeball data, as I reported last year, is a tough ask — even when free crypto is involved.)
Company pivot — So, faced with a scaling challenge, Tools for Humanity is now exploring alternatives to the Orb, its flagship product. “We’re going to be talking about, in addition to Orb-verified people, other methods, up and down the stack, to prove liveness and identity,” Traina tells TIME. Asked for further details, he demurs. “I don’t want to scoop us too soon,” he says.
Who to Know: John Coogan
John Coogan, the host of the popular Silicon Valley talk show TBPN, has waded into the Anthropic-OpenAI beef in the funniest possible way: by launching “Claude with Ads.” Regular readers of this newsletter will know that Anthropic fired shots at OpenAI in its Super Bowl ad over the weekend, by announcing that Claude would never have advertisements — a pointed criticism of OpenAI, which is piloting ads in ChatGPT.
But Coogan and TBPN seem to have calculated that if Anthropic won’t do it, they can do it themselves. Says Coogan on X: “I’ve wanted Claude to have an ad supported tier for years. Today, thanks to the Opus 4.6 API, Claude with Ads is here. Please enjoy intelligence too cheap to meter.”
Given that Anthropic has already proven itself happy to pull the plug on uses of its tools that it doesn’t like — like blocking OpenAI staff from using the Claude API — I’d wager that this website only has a matter of hours before it gets knocked offline. Enjoy the trolling while it lasts.
AI in Action
ChatGPT is back to “exceeding 10% monthly growth,” Sam Altman told OpenAI employees in a Slack message on Friday, CNBC reported. If it holds, that rate of growth would put OpenAI on course to surpass a billion weekly ChatGPT users very soon. The OpenAI CEO also said that the company was preparing to launch a new model this week.
What We’re Reading
The Internet’s New Favorite Insult: ‘Did AI Write That?’ by Angela Haupt in TIME
My colleague Angela Haupt writes about the newest insult on the block: accusing somebody of using AI in their writing: “Being told you sound like AI, then, can feel oddly dehumanizing. ‘That’s why the insult stings,’ Steele-Wren says. ‘It’s not about quality. It’s about identity. It suggests your voice is generic or interchangeable,’ and that hurts.”