Why Anthropic’s IPO is a Bet on Businesses Changing How They Work

What Happened

Anthropic’s proposed initial public offering has put the economics of the AI industry firmly under the spotlight. In a few years, we’ve gone from asking AI to generate text and images to using it to work through complex problems, support scientific research, and carry out tasks across systems.

Anthropic has now launched its pre-IPO roadshow, and part of the growth case it’s making to investors depends on the businesses that already use Claude finding more valuable things to do with it.

The prospectus devotes 93 pages — nearly one page in three — to risk factors. These cover a wide range of business and investment risks, but among them is an unusually stark warning: sufficiently advanced AI could pose “catastrophic or existential risks to humanity” if developed or deployed without adequate safeguards.

The prospectus also describes a problem much closer to the day-to-day work of an enterprise: an agent can take a series of actions at a speed and volume that make it difficult for the people overseeing it to intervene in time.

Read those parts of the prospectus together, and you get a sense of what Anthropic is asking of its customers. It needs them to give AI a bigger role in their businesses. And those businesses have to work out what they’re prepared to hand over, and how they’ll make use of the capabilities they’re buying in a productive (and safe) way.

Why It Happened

Central to Anthropic’s ambition is enterprise adoption, and the true economic value of increasingly intelligent AI will ultimately rely on what businesses can achieve with it.

Anthropic can keep improving its models, but its customers still have to decide how much to invest in changing the way their people work.

We’ve spent the past several years asking how intelligent AI can become. The industry has made extraordinary investments in response, but the more consequential point for business leaders now is how much more intelligent their organizations can become because of it. Anthropic can keep improving its models, but its customers still have to decide how much to invest in changing the way their people work.

Take an AI agent introduced to help employees resolve complex problems. It may be technically capable of doing the work. However, the employee still needs to know when to trust its response, when to question it, and what to do if it gets something wrong. If the same problem keeps coming up, someone needs to be able to see that pattern and do something about it. What the organization learns from those interactions can be just as valuable as the work the agent does.

That’s what I mean when I talk about a more intelligent organization.

The knowledge gained in one part of the business can change a decision somewhere else. Getting there means looking at how knowledge is shared, who has the authority to act on it, and where human judgment makes the most difference.

Those are choices about how the business actually runs, and they need funding and attention of their own.

What Happens Next

It’s inevitable that the pressure to demonstrate a return on AI will build. As it does, and from a leadership lens, I’d look beyond adoption numbers and ask what has actually improved. Are employees making better decisions? Are customers getting a better experience? If a team can now solve a problem it couldn’t before, can the rest of the organization learn from how that happened?

I’d apply the same scrutiny to what could get worse. That means thinking through the unintended consequences for employees and customers...

I’d apply the same scrutiny to what could get worse. That means thinking through the unintended consequences for employees and customers before AI takes on more responsibility. If you expect someone to intervene when a system makes a bad call, have you given them enough context to recognize it — and the authority to step in? That decision belongs in the design of the work from the get-go.

Anthropic is asking investors to put money behind what AI could become. But the companies buying it have a different decision to make. Before they sign off on the next round of AI spending, I’d ask them to put two things in the same business case: What do they expect the technology to do? And what are they prepared to change about the way they work so their people can use their expertise where it makes the most difference? Fundamentally, both need a budget.

Simply adopting sophisticated AI tools is insufficient; true business value and the next economic unlock will only come when companies redesign how they operate to effectively absorb and leverage the technology.

Keri Dawson is CEO and Global Head of Designit, Wipro’s experience innovation company.

Featured image by Luis Villa del Campo from Madrid, Spain, CC BY 2.0, via Wikimedia Commons . Creative Commons Attribution 2.0 Generic license.