The Sydney Morning Herald
· collected 2026-09-18 · by David Streitfeld
“If you’re explaining, you’re losing,” Ronald Reagan is credited with saying.
asserted
Reagan → explain → ?
Silicon Valley disagrees.
asserted
Valley → disagree → ?
Ever since the drama around artificial intelligence began heating up, technologists have unleashed bushels of commentary.
asserted
technologists → begin → commentary
Mark Zuckerberg, Meta’s CEO, wrote 6500 words about how people who use a lot of AI will survive best in the AI world to come.
asserted
who → write → world
Bill Gates spent nearly 6000 words arguing that AI will be humanity’s undoing unless we reorient human society and behaviour around it.
asserted
we → spend → it
Sam Altman of OpenAI, whose San Francisco house was attacked, needed only 1000 words to stress that “the world deserves huge amounts of AI and we must figure out how to make it happen”.
asserted
it → attack → AI
He has contributed to dozens of scholarly papers and more recently wrote a half-dozen informal essays for the general public.
asserted
He → contribute → public
Outside of a few references to such cultural artefacts as The Lord of the Rings, Carl Sagan and 2001: A Space Odyssey, these essays are generally measured, colourless, equipped with footnotes — an engineer’s reflections.
asserted
essays → equip → footnotes
Putting words into sentences and sentences into paragraphs is an old-fashioned thing to do, particularly when the topic being explained is a technology so amazing it might as well be magic.
uncertain
it → put → paragraphs
But as Anthropic moves ahead with what may be the largest public offering in history as soon as this fall, Amodei needs to forcefully reassure everyone that everything is on track.
uncertain
everything → move → track
He’s losing if he’s not explaining.
asserted
he → lose → ?
His collected essays are a chronicle of Silicon Valley’s constantly shifting methods of selling its own product to a public that is simultaneously enticed and horrified.
asserted
that → collect → public
It’s not an easy task, convincing the world that something is a slam-dunk even though it will make Silicon Valley richer and more powerful while possibly leaving ordinary folks unemployed or — who knows? — dead.
uncertain
who → ’ → folks
Amodei’s own explanations have changed with the political weather.
asserted
explanations → change → weather
In January, he wrote in a 20,000-word essay, The Adolescence of Technology, about his wariness of regulations, saying they backfire and quickly become out of date.
asserted
they → write → date
Even worse: “They can potentially destroy economic value or coerce unwilling actors who are sceptical of these risks (and there is some chance they are right!).”
asserted
they → destroy → risks
But in June, after the Trump administration asked AI companies to voluntarily submit new models for approval, Amodei said he was “grateful”.
asserted
he → ask → approval
New AI products were like airplanes, he wrote in a new essay: “Their release should be blocked or reversed as a threat to public safety if they do not meet high standards of safety.”
asserted
they → write → safety
Last week, the AI disaster narrative again seized the upper hand.
asserted
narrative → seize → hand
Jacob Coxon, an Anthropic employee, quit with the dire public warning that Anthropic and OpenAI were “gambling with our lives”.
asserted
Anthropic → quit → lives
Amodei responded with a 3800-word essay, We Must Pace the Frontier.
asserted
We → respond → Frontier
He proposed steps that would ensure AI companies like Anthropic slow down: “Pacing does not mean halting model training or technical progress, but ensuring companies take adequate time to align and safeguard their models, and for third-party evaluators to confirm this.”
asserted
evaluators → propose → this
An Anthropic spokesperson wrote in an email that “Anthropic/Dario remain very much in favour of regulation” but “in the absence of regulation”, the company was immediately implementing self-regulation.
asserted
company → write → regulation
Amodei’s most basic point is that AI is inexorable but that it is also extremely difficult — or maybe impossible — to control.
asserted
it → control → ?
Even before Anthropic existed, when he was working at Google in 2016, he co-wrote a paper titled Concrete Problems in AI Safety.
asserted
he → exist → Safety
It illustrates the many ways things can go wrong, using the example of a cleaning robot: “An office might contain pets that the robot, never having seen before, attempts to wash with soap, leading to predictably bad results.
uncertain
robot → illustrate → results
The growth in AI’s complexity since then has left the cleaning robot far behind.
asserted
growth → leave → robot
In The Urgency of Interpretability in April 2025, Amodei wrote: “We can’t stop the bus, but we can steer it.”
asserted
we → write → it
By January, in The Adolescence of Technology, he was warning: “I believe we are entering a rite of passage, both turbulent and inevitable, which will test who we are as a species.”
(The New York Times has sued OpenAI and Microsoft, claiming copyright infringement of news content related to AI systems.
asserted
Times → warn → systems
The two companies have denied those claims.)
uncertain
companies → deny → claims
Amodei’s most revealing essay by far, written in October 2024, makes the definitive case for Silicon Valley.
asserted
essay → write → Valley
Forget about the risks of AI blowing up, he wrote.
asserted
he → forget → risks
What will Utopia look like?
asserted
Utopia → look → What
The essay is titled Machines of Loving Grace.
asserted
essay → title → Grace
The source of the phrase is not acknowledged or explained anywhere in its 20,000 words, although there is a link in a footnote.
asserted
source → acknowledge → footnote
The title comes from a 1967 poem by Richard Brautigan, the poet laureate of the hippies, which ends like this:
I like to think
(it has to be!)
of a cybernetic ecology
where we are free of our labours
and joined back to nature,
returned to our mammal
brothers and sisters
and all watched over
by machines of loving grace.
Literary critics debate whether Brautigan was being sincere or tongue in cheek, a hallmark of the ambiguity that surrounds good art.
asserted
that → come → art
He wrote the poem during a stint at the California Institute of Technology, where he also wrote a brief poem that begins, “I don’t care how God-damn smart / these guys are: I’m bored.”
asserted
I → write → poem
Brautigan killed himself in 1984.
asserted
Brautigan → kill → 1984
There’s no question that Amodei is being sincere.
asserted
Amodei → ’ → ?
In Machines of Loving Grace, he wants people to join the cause because the results, even in the short term, will be so excellent.
asserted
results → love → term
…and 32 more, not listed.
The Aporia
