Damn, the doomers will not like it. Looks more like Eroom's law than Foom. We will see if this holds up.
Updated version!View quoted post on X
Latest posts 5
Yet another promising Parkinson treatment that made the disease worse(!) in a large phase II/III trial. Iron chelators failed. Now NAD failed too!
1/ Could AI predict such failures? I really doubt it. We can eke out some gains but it will not realistically solve the phase III problem.
2/ We need to run more trials and "fail faster". We need cheaper labor to do so and automation from AGI will be beneficial.
jamanetwork.com/journals/jama/…
1/ Could AI predict such failures? I really doubt it. We can eke out some gains but it will not realistically solve the phase III problem.
2/ We need to run more trials and "fail faster". We need cheaper labor to do so and automation from AGI will be beneficial.
jamanetwork.com/journals/jama/…
jamanetwork.com Nicotinamide Riboside for Patients With Early-Stage Parkinson Disease This randomized trial assesses the effect of oral nicotinamide riboside treatment for 52 weeks vs placebo on clinical outcomes among patients with early Parkinson disease. Daily reminder that AIxBio is already saving lives
AI assisted drugs are more successful than average
pubmed.ncbi.nlm.nih.gov/38692505/
Alphafold wins the Nobel prize
nobelprize.org/prizes/chemist…
FDA list of AI enabled devices
fda.gov/medical-device…
(several of which have strong RCT support)
AI assisted drugs are more successful than average
pubmed.ncbi.nlm.nih.gov/38692505/
Alphafold wins the Nobel prize
nobelprize.org/prizes/chemist…
FDA list of AI enabled devices
fda.gov/medical-device…
(several of which have strong RCT support)
pubmed.ncbi.nlm.nih.gov How successful are AI-discovered drugs in clinical trials? A first analysis and emerging lessons -... AI techniques are making inroads into the field of drug discovery. As a result, a growing number of drugs and vaccines have been discovered using AI. However, questions remain about the success of... The new AA data is interesting. It seems that as always google has a big model but the post-training just isn't there. High scores on omniscience but low scores on full and diverse AA index. Ant and OpenAI in a league of their own.
