Portuguese 
English 
Spanish 
3 min of reading 
0 comments 
An artificial intelligence framework developed at Peking University independently solved an open problem in commutative algebra, proposed more than a decade ago by American mathematician Dan Anderson, in just 80 hours. According to a report by the South China Morning Post, the work was released as a preprint on April 4, 2026.
The novel aspect is the level of autonomy. Previous systems required human supervision at some stage. This framework, according to the paper, essentially dispensed with any researcher intervention to solve and prove the result.
The problem had been formulated in 2014 by Dan Anderson, former professor at the University of Iowa, who died in 2022 at the age of 73. It belongs to the field of commutative algebra, a branch that studies rings and ideals used in algebraic geometry and number theory.
-
Second installment of the INSS 13th salary: check the payment dates and who receives
-
China accelerates investments in Bahia, connecting BYD electric cars, critical minerals, renewable energy, and the automotive industry in Camaçari, and places the Northeast at the center of a billion-dollar economic turnaround linked to clean technology in Brazil.
-
The idea is that giant spheres floating in the ocean generate energy from the movement of the waves, power AI chips inside them, and use seawater for cooling. This is the plan of Panthalassa, which has already raised US$ 200 million.
-
Embraer starts 2026 with R$ 7.6 billion in the best first quarter in history, accelerates deliveries of 44 aircraft in São José dos Campos, and stacks the 6th consecutive record in its backlog.
How Chinese AI solved the problem without human help
The architecture is a dual-agent system. One agent performs natural language reasoning about the problem. The other is responsible for formal verification, i.e., proving the result in rigorous mathematical language.
This design separates two abilities that, until recently, required human researchers. As a result, the AI can propose a path, draft the argument, and then check its validity line by line. Both agents work in parallel.
According to TechRadar, the system synthesized decades of mathematical literature to find the key to the problem. It cross-referenced ideas from different papers, generated new internal conjectures, and discarded those that did not pass the formal verifier.
Why this shakes the foundations of scientific research
Mathematics is the stronghold of proof. Everything needs to be formally correct. For decades, AIs only helped with searches and formatting. When they start contributing with autonomous proofs, the role of the mathematician changes.
Indeed, most of the world’s mathematics is done in cycles of years. An open problem for 12 years, solved in 80 hours, shortens the cycle by orders of magnitude. In other words, the frontier of what is considered “difficult” shifts.
On the other hand, the community will want to audit the result carefully. Proofs written by AI need to be reviewed by human mathematicians before they become accepted. According to a report by the Bangkok Post, this step is still underway.
The effect on the energy and infrastructure industry
It may seem distant, but commutative algebra is not an isolated world. Methods from the field appear in cryptography, logistics optimization, smart grid control, and algebraic geometry applied to engineering. Mathematical tools pave the way for technology.
Despite this, the direct impact will not arrive tomorrow. What will arrive is the possibility for AIs to assist engineers in complex calculations of operational risk, exploration models, and network planning. It is the next generation of professional software.
Ultimately, the Chinese artificial intelligence that solved Anderson’s problem raises a political question. Whoever masters these tools will also master the speed of discovery. However, Brazil still does not have a national laboratory dedicated to autonomous mathematical AI — and the time to enter this race is shrinking.
Be the first to react!