Line of inquiry
Inquiring lines›How do we keep AI systems safe and…›How does AI reshape human understa…›this line of inquiry
Can AI systems discover fundamental improvements to their own architectures?
A broader line of inquiry — a family of 33 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.
Questions in this line of inquiry 33
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Can bilevel autoresearch discover new search mechanisms for the inner research loop?
- Can bilevel autoresearch autonomously modify its own learning algorithms?
- Can AI systems learn their own objectives through autoresearch?
- How does automated mechanism discovery compare to human-led mechanistic research?
- Does discovering new AI architectures count as specified autoresearch or open-ended science?
- How does machine feedback enable discovery at test time?
- What makes AI-discovered architectures reveal design principles invisible to humans?
- What scaling laws govern autonomous architecture discovery in AI systems?
- Do models need world models to reliably steer hypothesis discovery?
- How does semantic search over research papers guide autonomous architecture proposals?
- How does executable evaluation feedback sustain autonomous discovery at scale?
- Can humans fully understand why the AI search strategy succeeded here?
- Does AIDE2's single loop differ from bilevel autoresearch's nested loops?
- How would a bi-level agent restructure objective functions during discovery?
- Why do major AI breakthroughs require human-discovered data and method combinations?
- What computational methods most reliably establish causal evidence in AI mechanism discovery?
- What interpretability challenges arise when algorithms are discovered rather than designed?
- What test-time strategies did o3 discover without human specification?
- Can human-aware models identify scientifically promising alien hypotheses reliably?
- How does bilevel autoresearch balance outer loop cost against discovery improvements?
- Do bounded awareness frames explain why AI optimization differs from open-ended discovery?
- What distinguishes a computational success like AlphaFold from a conceptual breakthrough?
- Can bilevel autoresearch succeed when the inner and outer loops use different models?
- Can wet-lab discovery remain autonomous when experiments require human hands at the bench?
- What distinguishes intrinsic search from extrinsic search method approaches?
- How many particles and iterations does optimal expert discovery require?
- Why do good algorithms become rarer as the search space grows more generic?
- How does the outer loop escape its own LLM's knowledge boundaries when discovering mechanisms?
- Can a single dominant mechanism replace the combined effect of all five?
- How much does domain shift limit the mechanisms a bilevel system can autonomously discover?
- Does Amdahl's law or partial substitutability better model research task automation?
- Why do automation waves follow the same pattern across different fields?
- What makes ripasudil's discovery for dry AMD a valid test of Robin's approach?