We may be in for a future we cannot yet imagine
On Sept. 29, Elon Musk stood outside the White House alongside President Trump and top artificial intelligence executives after an indoor AI leadership luncheon.
Addressing reporters on the driveway, Mr. Musk offered an optimistic prediction about where society is headed: “I think by far the most likely outcome is an age of abundance,” he said, describing a future in which advances in AI and robotics dramatically increase wealth and improve healthcare and living standards.
Such predictions can sound overly optimistic. For most Americans, AI is still largely associated with comparatively mundane tasks. People regularly use AI to summarize documents, write computer code, generate images and videos and help answer general questions.
Considering that the modern AI revolution is only a few years old, even these capabilities are remarkable.
AI is poised to transform virtually every aspect of our lives, including education, healthcare, logistics and advanced manufacturing. For the most part, AI functions as a force multiplier of human effort. It allows people to work faster and more efficiently, automating repetitive tasks and accomplishing more with fewer resources.
Yet something bigger seems to be on the horizon.
Recent reports say AI can solve problems, identify patterns and make discoveries at the outer edges of human understanding. It may be beginning to expand the frontier of human knowledge.
Tackling the unsolvable
On Sept. 8, OpenAI announced that one of its internal AI systems had produced a proposed solution to the Navier-Stokes problem, one of the famous Millennium Prize problems announced in 2000 by the Clay Mathematics Institute.
OpenAI turned a fleet of up to 10,000 parallel AI agents on the problem to see what they could come up with. The agents explored different approaches, shared useful findings and built upon one another’s work. After 88 hours, the system arrived at its proposed solution.
The answer will be analyzed by experts, but what happened was remarkable: a group of autonomous AI agents worked through this millennium problem in a matter of days.
Taking on biology
Just a few weeks after news broke about the Navier-Stokes problem, a similar report emerged in biology.
Anthropic researchers gave their AI system relatively broad instructions to comb through a massive DNA database and look for any unusual enzymes worth investigating. Nearly 1,000 AI agents went to work.
In 21 hours, the agents scoured a database of more than 200,000 reverse transcriptases, identified thousands of potential candidates and narrowed them down to the most interesting. One agent noticed an unusual repeating pattern in DNA sitting alongside a strange-looking enzyme.
After additional analysis and testing by scientists, Anthropic concluded that Claude, Anthropic’s AI model, had identified a previously uncharacterized enzyme system with features reminiscent of CRISPR.
Researchers still do not know exactly what this system does, but the significance lies in the method of discovery. AI appears to have noticed a biological pattern that humans have previously overlooked.
Read the rest at The Washington Times.
Donald Kendal is Director of The Heartland Institute’s Socialism Research Center. Kendal is also the host of Heartland’s In the Tank Podcast.
