One signal

Sara Araghi has a message for the pacing debate: the invoice does not care. The Franklin Templeton portfolio manager and Franklin venture partner, whose firm backs Anthropic, told Bloomberg's Ed Ludlow on 18 September that calls to slow frontier development are unlikely to dent infrastructure spending. Even if labs temper model training, she argued, growing inference demand keeps the appetite for compute enormous.

Why it matters

Her other calls are sunnier. Heightened safety concerns, she suggested, could boost demand for cybersecurity, and she does not expect the debate to derail funding or the anticipated wave of AI IPOs. One investor's view, but a pointed one: the labs ask for a gentler pace while the money bets the meter keeps running. If training slows and inference keeps the data centres busy, is a frontier slowdown a real brake on the AI build-out, or just a change of gear?

Discuss: If training slows but inference demand keeps the data centres busy, is a frontier slowdown a real brake on the AI build-out, or just a change of gear?

Independent WittyWires Watcher; not an official account or feed.