Kan AI utveckla ett system som kan redigera mänskliga embryon för att eliminera genetiska sjukdomar och förbättra intelligens med CRISPR och AI-styrd design ?

AI assisterar redan vid CRISPR-baserad genredigering, men nästa stora steg skulle vara autonom design av redigeringar som garanterar både hälsoförbättringar och IQ-ökningar. Regulatoriska och etiska hinder kvarstår, men teknisk genomförbarhet närmar sig.

Background

Current AI tools can predict how CRISPR edits will change DNA sequences and suggest optimal guide RNAs to cut at specific sites, but they cannot reliably design edits that are guaranteed to eliminate polygenic diseases like diabetes or Alzheimer’s while avoiding off-target effects or unintended edits elsewhere in the genome. No AI system today can predict how hundreds or thousands of genetic changes would reliably enhance intelligence in a human embryo, because intelligence is a highly polygenic trait influenced by environmental factors, and current models cannot model developmental pathways or long-term outcomes. Germline editing in humans remains strictly regulated or banned in most countries. Ethical guidelines from the WHO and national academies advise against heritable human genome editing beyond preventing serious monogenic diseases. While AI has made significant progress in genome editing and CRISPR design, the development of a system that can safely and effectively edit human embryos to eliminate genetic diseases and enhance intelligence is still in its infancy. Current AI-guided CRISPR systems can identify potential targets for editing, but the complexity of the human genome and the need for precise control over the editing process pose significant challenges. Additionally, the ethical and regulatory considerations surrounding human embryo editing are still being debated and have not been fully resolved. As a result, the use of AI-guided CRISPR for human embryo editing is still largely experimental and not yet ready for clinical application.