Lovable CEO Anton Osika reached for an old C.S. Lewis essay to make a modern point: people should not let AI fears swallow ordinary life.
In an X post on Saturday, Osika said, 'Every unprecedented moment has a precedent somewhere.' He cited Lewis' 1948 essay 'On Living in an Atomic Age,' which argues that people facing catastrophe should keep doing 'sensible and human things' instead of freezing in fear.
Lewis wrote that if an atomic bomb comes, it should find people 'praying, working, teaching, reading, listening to music, bathing the children, playing tennis, chatting with friends over a pint and a game of darts' rather than 'huddled together like frightened sheep and thinking about bombs.' He also wrote that bombs 'may break our bodies,' but they 'need not dominate our minds.'
Osika's post followed a sharper warning from Jacob Coxon, a 27-year-old researcher who spent about three years at OpenAI and Anthropic before quitting the latter. Coxon said the two frontier labs were racing toward self-improving superintelligence while 'gambling with our lives.'
Osika said on LinkedIn last week that warnings like Coxon's 'need to be taken seriously.' He added that the companies building the most powerful AI systems should have to prove they can keep them under control, and he backed slowing dangerous development when safety cannot be demonstrated.
Not everyone sees an extinction-level threat. AI researcher Gary Marcus said there was 'no realistic scenario' for near-term human extinction, while Santa Fe Institute professor Melanie Mitchell called such fears 'far-fetched.'
Over the weekend, Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman also backed efforts to pace frontier AI development and bring in third-party safety evaluators.
The policy backdrop matters for founders. The Trump administration has shown little appetite for regulating the sector as it seeks to maintain the US lead over China. Trump said on Sunday that 'We're leading China in AI, we're the most sophisticated country in the world,' and added, 'whoever wins AI, wins.'
Here's the playbook for founders: the race for AI leadership is not slowing just because the warnings are louder. That puts the burden on builders to prove reliability, safety, and control before scale outruns trust.
The math is simple. In a market where frontier models can move faster than rules, the winners will be the companies that can ship fast without making customers, investors, or regulators question whether the system is safe to use.
