Anthropic surges as OpenAI struggles to keep up

Hello, and welcome to TechScape. I’m your host, Blake Montgomery, US tech editor at the Guardian. This week in tech, we’re discussing Anthropic’s meteoric rise, both theological and financial, and California’s unprecedented infusion of political cash from Silicon Valley.

Anthropic, the AI firm behind the Claude chatbot, announced on Monday it had filed confidentially for an initial public offering on the US stock market. The company announced the filing in a short, two-paragraph blogpost that did not give an exact timeline for the company to go public and did not reveal the number of shares that it would offer.

The filing continues the company’s banner year, as well as pre-empts its rival OpenAI, which is expected to imminently file for its own IPO. It’s the second blow in the rivalry in just a week. Las week, Anthropic leapfrogged OpenAI to become the world’s most valuable startup when it announced it had raised $65bn in funding to value the company at $965bn, eclipsing OpenAI’s $852bn valuation. Anthropic was valued at $380bn in February, but, as the Wall Street Journal reported, the company’s “mind-blowing” revenue growth, particularly thanks to its coding tool that has proved extremely popular with business clients, is poised to bring the company its first profitable quarter in June.

Once a smaller player in the AI race, Anthropic’s rapid ascent over the past year has placed it neck-and-neck with OpenAI in terms of dominance over the industry. The filing is a triumph in that ascension. OpenAI, where Anthropic’s CEO, Dario Amodei, once worked, has struggled to keep up. OpenAI also operates a coding product, Codex, which is not nearly as popular as Claude Code. Anthropic’s Claude Mythos, a cybersecurity AI bot, made a major international splash in April with its ability to find bugs in widely used software. OpenAI’s similar product, released weeks later to little fanfare, looked like an also-ran.

The reversal of fortune for ChatGPT’s maker is a remarkable loss of first-mover advantage, which may affect investor appetite. SpaceX, Anthropic and OpenAI are all slated to debut on the stock market this year, inflating it by at least three trillion dollars, and the level of hype surrounding each company will determine its valuation. If OpenAI continues to lose ground to Anthropic, it seems likely that Sam Altman will become the veep to Amodei’s president.

‘Like a billionaire on acid’: Star Wars director Gareth Edwards comes out in favour of AI

To YouTube and beyond: how online gen Z directors stormed Hollywood

In an encyclical – the first major text on safeguarding humankind of his papacy – Pope Leo denounced the “culture of power” driving the rapid rise of artificial intelligence, while warning that the technology must be subject to the “most rigorous” ethical constraints as it infiltrates everything from work to war. The pontiff delineated the technology’s most concerning threats to humanity: replacing workers, accelerating war and exploiting the environment.

Beside the pontiff was a co-founder of Anthropic, Chris Olah, one of the people behind the AI boom so worrying Leo.

Leo’s address inspired both skepticism and adoration. My colleague Sanya Mansoor asks: why did Anthropic’s founder sit with the pope during a warning about AI?

Olah’s presence raises a key question: how could the Catholic church and the world’s most valuable AI startup work together, when Anthropic’s technology may bring about the future Leo is warning against?

“All of these companies are building technology that … is designed to replace people,” says Pete Furlong, senior manager of policy and research at Center for Humane Technology, a non-profit advocating for accountability around AI. “That’s very much at odds with the pope’s words. You can’t have dignity in a world where you’re building technology to replace people.”

Timnit Gebru, founder of the Distributed Artificial Intelligence Research Institute, wrote in a Linkedin post that the alliance was effectively “Vatican-washing” and said the church should have partnered with “the exploited data workers fighting for their rights, the people whose water is polluted fighting datacenters, or the many other victims around the world”.

Olah’s proximity to Leo is a canny alliance for both Anthropic and the pope. Magnifica Humanitas, the title of Leo’s text, translates to “Magnificent Humanity”, which relates to Anthropic’s name, meaning relating to humans. The Vatican and the startup proclaimed a shared emphasis on humanity.

The alliance also makes for a savvy bit of marketing. The company burnishes its brand of safety-first AI by sitting alongside a prominent leader advocating for the thoughtful development of artificial intelligence, ergo for the de-acceleration of the current unconsidered pace. (Even though Anthropic seems to have no intention of slowing down its schedule of product releases.) Leo, on the other hand, allies himself with the startup that, as of this week, seems to have pulled into first place in the AI race. He positions himself as a critic on the cutting edge, sharing Anthropic’s power.

Francine Prose, a former president of PEN American Center, described why she found Pope Leo’s treatise powerful:

The letter says the most important and necessary things about what is possibly the greatest threat posed by AI: it can be programmed solely to maximize profit, a situation that can only result in the suffering of the many for the benefit of the few.

The encyclical’s vision of human nature, of the spirit of justice and empathy that needs to prevail, of the essential importance of the highest moral values – is ultimately so beneficent, so positive, so generous, so inarguably clear about our obligation to protect the weak and the poor that it’s hard to find reasons to dismiss it.

If the masters of this new technology fail to agree with what the pope sees as its dangers and drawbacks, we are in very deep trouble indeed.

Both the critical response and the laudatory one have more in common than might seem. Both express a wariness of tech’s oligarchs, though to varying degrees. The “Vatican-washing” response demands a more severe excoriation of AI from Leo; Anthropic is the villain. The response of praise shows gratitude for Leo setting himself against Silicon Valley’s move-fast-and-break-things modus operandi; Anthropic is the ally.

Silicon Valley’s billionaires have made their home state’s primary elections, in which voters decide which candidates will appear on the ballot in November, into the costliest primaries in California history. The goal of such major spending, experts say, is to gain both political and regulatory leverage that will perpetuate dominance in business.