one year on
Future of Life Institute statement calls for prohibition on superintelligence development
A cross-spectrum coalition of AI scientists, celebrities, and national security figures says no lab should build superintelligence until it can be proven safe, controllable, and publicly endorsed.
Originally published Published on this site
The Future of Life Institute on Wednesday published a statement calling for a prohibition on the development of superintelligence—AI that could exceed the intelligence of all of humanity—until there is broad scientific consensus that it can be built safely and controllably, and strong public buy-in.
The signatories number more than 1,000 and span an unusual political and cultural range: AI pioneers Geoffrey Hinton, Yoshua Bengio, and Stuart Russell; Apple cofounder Steve Wozniak; Virgin founder Richard Branson; Prince Harry and Meghan, Duke and Duchess of Sussex; actor Joseph Gordon-Levitt; musician will.i.am; former Joint Chiefs chairman Mike Mullen; and Trump ally Steve Bannon. The statement argues that the pursuit of superintelligence raises risks of economic displacement, disempowerment, and threats to national security and civil liberties, and accuses tech companies of chasing the technology without guardrails, oversight, or broad public consent.
Polling released alongside the statement by FLI found that 64% of U.S. adults agreed superintelligence should not be developed until it is provably safe and controllable, while 73% supported robust regulation of advanced AI. Only 5% backed the current status quo of unregulated development.
Notably absent from the signatories: leaders of the labs most often associated with frontier AI. Leaders of OpenAI, Google DeepMind, and Meta are not among the signatories.
The record
said that 95% of Americans do not want a race to superintelligence, and that experts want to ban it
said that frontier AI systems could surpass most individuals across most cognitive tasks within just a few years, and that to safely advance we must scientifically determine how to design AI systems fundamentally incapable of harming people
told NBC News that the path has been chosen for us by AI companies and founders and the economic system driving them, but that no one has really asked almost anybody else if this is what we want
said in a statement that to get the most from what AI offers, there is no need to reach for superintelligence, which would result in a power we can neither understand nor control
One year later — open only if you can handle spoilers
The statement kept gathering names, about 75,000 by September 2026 including 5,000 collected through an Ekō petition, while the compute race it targeted sped up: Anthropic signed a gigawatt-scale TPU deal the next day and a multi-gigawatt expansion in April 2026.
Filed
Compiled and edited by Adam Allcock.
The Weekly Replay · free by email
This week, one year ago — every Sunday.
One email each Sunday: the week's replayed AI news, with the one-year-later annotations included. Written like it's breaking — dated like it isn't.
Free · double opt-in · unsubscribe anytime · privacy