Gary Marcus warns of infinite bot hackers, not rogue superintelligence
AI commentator Gary Marcus spent Sep 18-19 posting and reposting the same warning: the real near-term danger is not a sci-fi "runaway superintelligence" but a disaster that is already largely underway—with enough money, one can direct unlimited bots to continuously attack any server using every publicly known vulnerability until they succeed.
Confirmed
- These views were posted and repeatedly amplified by Marcus himself on @GaryMarcus, concentrated on Sep 18-19.
- He argued such mass automated attacks require no superintelligence or self-improvement and are "already largely happening."
- In a Mashable interview, he said people anthropomorphize basic cybersecurity lapses and get swept into hype, diverting attention toward fantasy "AI extinction" scenarios.
Why it matters
- Marcus argues the urgent question is what practical steps can protect the world's critical infrastructure from malicious AI use, rather than indulging distant extinction narratives.
- His repeated amplification points to the real gap between current agentic AI capabilities and cyber defenses, with direct implications for AI safety policy debates.
2026-09-18 ~ 2026-09-19 · 5 related posts
Primary sources
- [source] Gary Marcus: fixation on AI extinction scenarios distracts from practical misuse defenses — GaryMarcus · 2026-09-18
- [source] Gary Marcus: The near-term AI disaster to fear is infinite bot-driven hacking, not rogue AI — GaryMarcus · 2026-09-18
- Gary Marcus: The near-term AI risk is agentic AI hacking the internet at scale, not rogue superintelligence — asusarla · 2026-09-19
2 near-duplicate retellings: GaryMarcus · GaryMarcus