Artificial intelligence (AI), synthetic biology, and gene editing are powerful technologies that are being developed with remarkable speed and are becoming part of the economic and social landscape. The challenges of regulating their use and the industries they promote is increasingly urgent, not just for individuals, not only for nations or regions, not even for the human species – but also for the planet and all its inhabitants, animal, vegetable and mineral. Issues in our world don’t get any larger than that.
Most of the attention in recent years has focused on AI. Some of the loudest warnings about the dangers ahead, as well as the potential benefits, have come from exceedingly wealthy technologists – multi-millionaires, billionaires, even centibillionaires – who clearly think that they should make the decisions as well as the profit. They have gathered in substantial groups, notably at a 2015 conference in Puerto Rico and a 2017 event at Asilomar (in Big Sur, California), the site of the by-now legendary 1975 meeting focused on recombinant DNA. The 1975 meeting essentially jump-started the biotech industry; the 2017 one developed “one of the earliest and most influential sets of AI governance principles.”
Mustafa Suleyman, the lead author of The Coming Wave, was on the organizing committee of the first AI conference and a panelist at the second. Neither meeting features in the book; perhaps they are now considered of historical interest only. Suleyman has not needed to work for a living since 2014, when Google bought Deep Mind, a company he co-founded, for a reported $400 million. He carries on, however, and co-founded Inflection AI in 2022, which is now valued at $4 billion; Suleyman’s personal fortune is estimated at roughly half a billion dollars, maybe more. He was part of the small group that met with President Biden in July to discuss how to regulate the exploding AI industry. The Coming Wave is clearly intended as a contribution to that conversation. From the flap of the book jacket:
This groundbreaking book from the ultimate AI insider establishes “the containment problem”—the task of maintaining control over powerful technologies—as the essential challenge of our age.
That is exactly what makes Suleyman’s book valuable. Moreover, and of particular interest to those of us whose principal focus is abuse of human genetic technologies, the authors consistently call artificial intelligence, synthetic biology, and heritable human gene editing parts of the same “wave” that badly needs to be addressed — legally, socially, and internationally. This perspective, in a broad social and economic context, is potentially extremely useful.
Clearly, that is the intent of the authors and publisher. The book’s front cover has a quote from Yuval Noah Harari, and the back displays advance praise from a dozen well-known and distinguished people, including Bill Gates and Eric Lander. When you open it, before even the title page, there are six full pages of additional plaudits — extended versions of all the quotes on the back cover, and similar praise from another 16 public intellectuals. Many of them are featured on the website, along with a 12-minute video explainer, glossary, and bibliography. This is a concerted effort to spark and focus conversations, preferably on a global scale, at least in the anglophone world.
So, what is the problem with The Coming Wave’s approach? Actually there are several, among them dramatic under-estimation of technical difficulties, especially when combined with social issues, and wholly inadequate discussion of specifically political considerations.
First, there is the casual assumption that AI technology works, or very soon will work, consistently, accurately, and affordably. Perhaps this is true of some AI applications, but they certainly have limitations: AI routines may be superb at playing Go but not necessarily capable of understanding the creative ways in which a human can mess things up.
Suleyman breathes rarified air, as befits a rich man, but he has embarrassingly personal experience of messing up. DeepMind placed Suleyman on leave for, essentially, being a jerk to employees, and he shifted over to Google. There, by his own account, he worked on “experimental efforts to create innovative governance structures.” This involved in part establishing an external ethics advisory council, which fell apart immediately, because Kay Coles James, the conservative president of the Heritage Foundation, who had made “a number of anti-trans and anti-LGBTQ remarks,” was named as a member. More than 2,500 Google employees signed a petition calling for her removal, she refused, and the company canceled the project entirely, only nine days after it was announced. Suleyman writes that he personally disagreed with her opinions but argued that “the full range of values and perspectives deserved to be heard” since Google is a global company and some of their customers might share these prejudices. Profit over principle is never a good look.
All this seemed even worse after Timnit Gebru and Margaret Mitchell were “removed from their roles at the company” for complaining about the potential for Google’s technology to reproduce social biases. Gebru and other social activists working on AI issues, along with scholars such Ruha Benjamin, are conspicuous by their absence from The Coming Wave, which is a remarkable and important omission. Bias in AI has become an important topic at governmental levels and in academia, but gets short shrift here.
The Coming Wave blithely calls CRISPR “arguably the biggest biotech story of the twenty-first century” (p. 130). Certainly its discovery was a dramatic advance, as reflected by the Nobel Prize for three of its developers. But the much-touted advances of medical gene editing, for example, have not yet arrived, and it is by no means clear that affordable and widely available treatments will happen. And when it comes to genetic modification of human embryos — the inevitability of which this book assumes — the scientific practicality is very much in question (even setting aside ethical and social issues). Three years ago, Nature published an article headlined “CRISPR gene editing in human embryos wreaks chromosomal mayhem.” The mayhem noted at that time — “large DNA deletions and reshuffling” — has not been resolved to date.
The book does place considerable emphasis on the need for global regulation of both AI and gene editing, but rather hand-waves away the difficulties, apparently on the grounds that, well, they have to be overcome and therefore they will be. Which is not entirely convincing. Thus, on p. 265:
A good example here comes from germ-line gene editing. A study of 106 countries found that regulation of germ-line gene editing is patchy. … It doesn’t add up to a global framework on a technology with global scope. More effective to date is the international collaboration of scientists on the front line.
At least he noticed the 2020 analysis by Françoise Baylis, Marcy Darnovsky, Katie Hasson, and Timothy M. Krahn (cited in the endnotes). But does he mention the series of international conferences on heritable genome editing? Or the reports by the National Academies, the Royal Society and the World Health Organization? No, not once. To back up his optimism about the collaboration of scientists, he refers only to a 2019 paper in Nature, by Lander, Baylis, Zhang, Charpentier, Berg and 13 others (from in total 7 countries):
Adopt a moratorium on heritable genome editing
This call, Science helpfully noted at the time, “is a departure from statements or reports issued earlier by two global summits on genome editing.”
Introducing the book, Suleyman writes: “Both pursuing and not pursuing new technologies is, from here, fraught with risk.” He describes how tech CEOs at a meeting that he addressed were shockingly complacent about the future of technological development. He doesn’t need to comment that the modern overlords still chase the wealth of the founders of Tesla, Amazon, Microsoft and Google, some of the richest people in the world.
Given the current political and economic situation, it is far from a stretch to claim, as The Coming Wave does on page 254, that “Profit drives the coming wave. There’s no pathway to safety that doesn’t recognize and grapple with this fact.” The book does point out, near the end, that “private sector salaries can be ten times their public sector equivalents” and rightly calls this “unsustainable” but fails to note that nobody in the public sector has any chance at all of legally amassing from their entire career 10% or even 1% of the half-billion or so that Suleyman has racked up in less than 15 years.
The book does include a few suggestions on how to rein in runaway capitalism, but its contribution lies more in provoking discussion than in providing solutions. And Suleyman evokes, on p. 270, what he clearly sees as the inspirational spirit of 1975:
As the wave keeps building, we will need to self-consciously return again and again to the spirit — and letter — of Asilomar.
Ah, Asilomar! The source of the guidelines that fueled the biotechnology industry and made quite a few scientists millionaires, back in the days when that meant “very rich.”
So, why would I recommend this book? Essentially, know your enemy.