Earth to tech bros: is there any intelligent life in Silicon Valley?

In the mid-2000s, my Apple MacBook was looking for intelligent life beyond Earth, even when I wasn’t using it. I was part of the global project called SETI@home, one of many diverse efforts in the search for extraterrestrial intelligence (Seti). You downloaded some software from Berkeley and allowed it to run in background mode on your machine, doing signal analysis on data from an American radio telescope that was looking for intelligent signals from outer space. At its height, the project had 180,000 volunteers, thereby creating a kind of global supercomputer. Sadly, no promising signals were found.

We now need a contemporary project – a search for terrestrial intelligence (STI, perhaps) – that will look for signs of intelligent life in the torrent of hype and misinformation about AI emanating from Silicon Valley.

This is an industry, remember, that warns us that, although it’s creating technology that may constitute an existential threat to humanity, it should not be prevented from building it. AI, burbled one prominent venture capitalist, is “quite possibly the most important — and best — thing our civilisation has ever created”. Google’s chief executive thinks it “more profound than fire or electricity”. AI will, Elon Musk famously told former prime minister Rishi Sunak in a televised conversation, make work “optional”. It will, said Anthropic’s boss, compress a century of biomedical progress into five to 10 years and cure most diseases. At the same time, though, it could also eliminate half of entry-level white-collar jobs within years.

And, of course, all this is going to happen real soon now. It’ll be like a tsunami: unstoppable. And it moves too fast to be regulated by dozy governments, assuming that they have not already put up the white flags of surrender to this irresistible force. And so on, apparently ad infinitum.

What’s driving these narratives is the crazed logic of a stampede. It has two dimensions. One is geopolitical: the US is suffering hegemonic anxiety that mastery of the technology may pass to China. The other dimension is industrial. A small number of American tech companies are locked into a frenzied and financially disastrous arms race in the belief that “AI supremacy” is essential for their survival. In the process, they are blowing unconscionable amounts of capital without a business model that could credibly achieve the returns needed to justify that level of investment.

This is unsustainable, and when the bubble finally bursts, it will have serious repercussions. As of now, Apple, Microsoft, Alphabet, Amazon, Nvidia, Meta and Tesla make up about a third of the US stock market. If their valuations dropped by, say, 40%, then the S&P 500 itself would fall by about 13% or 14% purely from that one basket – even though 493 other companies didn’t move. And the managers of your pension fund (and mine) may begin to have sleepless nights.

The challenge for the rest of us is how to keep our heads about AI when all around us people are losing theirs. Here are a few suggestions.

First of all, denialism is no longer a strategy. This is a significant new general-purpose technology, like electricity. And, like electricity, in the long run, it will affect how we all live and work.

On the other hand, treat the deterministic narrative of the industry with the scepticism it deserves. It indicates that boosters know little of the history of earlier technological revolutions, and in particular have no idea of how long it takes to bring about big industrial change.

As the Princeton computer scientist Arvind Narayanan put it the other day: “There are so many bottlenecks between AI capability increasing and it having these impacts, both good and bad, especially if you want to really integrate it into the economy and derive something useful out of it. There are so many things that have to go right. We think those are going to happen over a period of decades, not months or years.”

It’s worth remembering that mainstream media coverage of the technology is very patchy and sometimes misleading. For example, it tends to focus on certain kinds of mundane or repetitive tasks where the use of a large language model is an obvious application. This kind of reporting may make for good copy, but it doesn’t foster public understanding.

Also, mainstream media have not focused much on the actual experience of companies and organisations that have tried the technology. Consequently, stories such as the release of recent research by MIT – revealing that 95% of the 52 organisations surveyed had achieved zero return on AI investment, despite spending between $30bn and $40bn on generative AI across more than 300 initiatives – aren’t widely reported. There’s a lot more in that vein you’ll never hear about from AI evangelists.

The Seti project needed my spare processor cycles. By contrast, STI just needs your spare scepticism – running quietly in the background – the next time someone tells you the tsunami is unstoppable, that regulation is futile, or that a century of medicine is about to be compressed into five years.

No promising signal has come back from Silicon Valley yet. But there is intelligent life on Earth: you just have to go looking for it – one dubious press release at a time.

Photograph by FlixPix via Alamy

Photograph by FlixPix via Alamy