models ace frontier math but can't count strawberries
openai published solutions to longstanding problems that shocked mathematicians, sparking a debate about the existential threat to the field.
openai recently published solutions to high-level mathematical problems, a feat that has caused a 'bombshell' debate within the academic community. while models still struggle with elementary tasks, their rapid improvement in abstract mathematics challenges the traditional role of human mathematicians in identifying and solving problems. this development raises concerns about the viability of academic training programs and whether frontier ai labs' interest in math is merely a marketing exercise rather than a genuine pursuit of knowledge.