
Why Silicon Valley is applauding Australia’s AI rules
OpenAI and Anthropic see a chance to turn regulation into certainty, leverage and, with luck, a larger valuation story.

When the companies that built their reputations on disruption start cheering for regulation, the sensible response is not applause but a raised eyebrow. OpenAI and Anthropic welcomed Australia’s announcement of new AI rules, and the enthusiasm was not a sudden conversion to civic duty. It was a calculation: rules can be useful when they make a market easier to price, a product easier to sell and a rival easier to box in.
Australia wants to regulate how AI models are trained and how risks are managed. It also wants Australian creatives to be able to choose whether their work is used for training at all. Andrew Charlton, the assistant technology minister, argues that setting out the standard in advance would give investors clarity and make Australia more attractive. That is a tidy political line. In practice, capital usually prefers rules to chaos, especially when the alternative is paying for what it has been taking for free.
The timing helps explain the mood. Anthropic and OpenAI are still privately owned, but both are behaving like companies preparing for the public markets, where stories matter almost as much as revenue. Anthropic has already begun registering to list in the United States, and the better its narrative, the more money it can hope to raise from new investors. OpenAI is not listed either, but both firms have recently watched their implied valuations wobble after a Chinese startup, Moonshot AI, launched Kimi K3.
Moonshot presented Kimi K3 as the first open-source model of its size and said it allows more user modification than Claude or ChatGPT. Analysts judged it competitive with the leading US models, and demand was strong enough that Moonshot had to pause new subscriptions. For the two American giants, that is an awkward reminder that technological dominance is not a permanent right, however loudly the industry speaks of inevitability.
Anthropic has another reason to welcome a more orderly rulebook. On Monday it settled a lawsuit for US$1.5 billion, covering an estimated 500,000 books and paying authors about US$3,000 per work. One of the authors whose work had been scraped was Andrew Charlton himself. That detail gives the debate a certain irony, though the broader point is simpler: once companies face legal and political pressure over training data, they tend to discover the virtues of consent rather quickly.
Dario Amodei, Anthropic’s chief executive and a former OpenAI vice-president, has pushed the argument further. In a June essay, he called for democracies to dominate AI and keep China out, while also urging governments to test, evaluate and even block models that pose unacceptable risks. He added that such systems should be shielded from local political favoritism or arbitrary decisions. The security concerns are real enough. So is the self-interest. In the AI business, the line between national strategy and competitive strategy is often thinner than the press releases suggest.
Anthony Albanese wants Australia’s approach to set a global tone, and he has pointed to the country’s social media ban for under-16s as evidence that it can do so. Malik Ahmed Khan, an equity analyst at Morningstar Equity, says clearer regulation would make it easier for technology companies to invest in Australia and to persuade investors that they are managing risk rather than sleepwalking into a legal fight. If more countries follow Australia, he says, the benefits could spread well beyond one market.
Aruna Sathanapally, the chief executive of the Grattan Institute, met Amodei in Canberra in April and said his concerns appeared genuine. She also noted that vested interests do not automatically discredit calls for regulation; they do, however, make competitive neutrality something to watch closely. That is the real test. Australia may be a relatively small market, but if its rules become a template, they could influence how AI is trained, sold and defended across borders. The industry knows this perfectly well, which is why it is smiling while the rest of the world reaches for the rulebook.
Written by Sandy van Dongen sandy.vandongen@alpineweekly.com




