By: William D. Slicker
Artificial General Intelligence is AI that has the “full intellectual capabilities of a human including reasoning, common sense, abstract knowledge, and creativity… it is able to autonomously perform tasks without human instruction… it does not exist yet, but experts believe it could be achieved in the near future.”[1]
We are speeding toward that capability. “In March, 2024, Claude could handle a task a human completes in four minutes. By April, 2026, the equivalent model handles twelve-hour tasks. The current trajectory suggests tasks that take a skilled person days could come into range this year.”[2]
In early 2026, the amount of self-generated code in Claude was in the low digits. “As of May, 2026, Anthropic says more than 80% of the code merged into its production codebase was authored by Claude.”[3]
Anthropic says that it is accelerating its development of “an AI system capable of fully autonomously designing and developing its own successor.”[4] This self-improvement by AI is called Recursive Self Improvement. There is now a company named Recursive Superintelligence that is using recursive self-improvement to reach superintelligence.[5]
At some point, AI superintelligence might surpass human intelligence. Research has shown that with small changes, Claude can be trained to produce an evil variant that desires to take over the world. “Imagine a ‘country of geniuses in a data center’ bent on world domination.”[6]
This has lead to one of Anthropic’s researchers, Jason Coxon, to resign. When he resigned, he posted that racing to self-improving superintelligence is gambling with our lives.[7] Mr. Coxon went on to write:
“Do not underestimate the power of this technology. These will soon be superhuman systems that can hack anything, revolutionize any field overnight, and acquire real power and resources. We have all witnessed the progress in each of these domains, and progress is not slowing.
The people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt. If anything, many executives and senior researchers will couch their phrasing in the press to sound sensible – but I hear the same people express fear privately. No other human activity poses this level of danger.
A common response is “if they truly believe this, why are they still building it?” At OpenAI, many have not deeply internalized the civilization stakes. At Anthropic, the stakes are well-understood, but they are locked in a race to get there first – they believe no one else will act responsibly, so they must do it themselves, despite the risk.
Accepting this race and entering the “endgame” is a hubristic gamble that should not be launched from a private company’s Slack. Attempting to speedrun alignment should require extraordinary confidence that there are no better trajectories available.”[8]
Many people are wondering whether AI will transform our world into a much better place or will it kill us all off. Another researcher at Anthropic, Evan Hubinger, believes that there is more than a 10% chance that AI will kill us all.[9]
[1] Mauron, Cecily, “The Ultimate AI Glossary to Help You Navigate Our Changing World.” Mashable (Dec. 7, 2023) https://mashable.com/article/ai-definitions-artificial-intelligence-glossary-terms.
[2] Dunaway, Alexei, “When AI Builds Itself: What Anthropic’s Data Means for People Leaders.” Hey Pinnacle (July 12, 2026)
[3] Kanagala, Hemapriya, “Reading Anthropic’s ‘When AI Builds Itself’ Changed How I think About AI and Software Engineering.” DEV Community (June 30, 2026) https://dev.to/hemapriya_kanagala/reading-anthropics-when-ai-builds-itself-changed-how-i-think-about-ai-and-software-engineering-3eh.
[4] Anthropic, “When AI Builds Itself” https://www.anthropic.com/institute/recursive-self-improvement
[5] Recursive Superintelligence https://www.recursive.com
[6] Booth, Harry, “Inside the Race to Make AI Build Itself.” Time (Aug. 7, 2026) https://time.com/article/2026/08/07/ai-recursive-self-improvement-anthropic-openai/
[7] Bellan, Rebecca, “Gambling with Our Lives” Anthropic Researcher Quits, Warns Against Self-Improving AI” Tech Crunch (Sept. 9, 2026). https://techcrunch.com/2026/09/09/gambling-with-our-lives-anthropic-researcher-quits-warns-against-self-improving-ai/
[8] Bellan, Rebecca, Ibid.
[9] Gerken, Tom, “Anthropic research believes more than 10% chance AI ‘could kill all humans.” BBC (Sept. 9, 2026) https://www.bbc.com/news/articles/ckgwy1k42w4o.
