Will AI will discover a new drug promising enough for clinical trials in 2022?

620Ṁ3085

resolved Jan 1

Resolved

YES

ALL

From Vox Future Perfect’s predictions for 2022: https://www.vox.com/future-perfect/22824620/predicting-midterms-covid-roe-wade-oscars-2022 Resolved according to Vox’s judgment, as given through future posts.

Science

Biotech

Get

1,000

to start trading!

🏅 Top traders

#	Name	Total profit
1		Ṁ412
2		Ṁ69
3		Ṁ55
4		Ṁ34
5		Ṁ27

People are also trading

Will there be an AI that diagnoses patients better than an average doctor prior to 2030?

68% chance

Will an AI-designed drug get FDA approval by the end of 2025?

19% chance

Will a drug designed with significant contribution from an AlphaFold-like system pass a Phase I trial by end of 2025?

21% chance

Will an OpenAI model design an improved version of an existing drug in 2025?

24% chance

Will AI developed molecules or drugs be the dominant sources of new biopharma IP by Sep 5 2029?

74% chance

Will AI be regularly used to diagnose complex diseases by 2028?

32% chance

Will FDA approve an “AI doctor” for primary care by 2025?

5% chance

Will the markov.bio AI copilot for drug discovery be influential in the R&D of a marketed drug? (Before 2030)

8% chance

Will an AI be able to write a prescription for medication in the US or a EU member country by 2030?

17% chance

The FDA will approve an AI developed cure before the end of the decade (<2030) for a previously incurable major illness.

Sort by:

https://www.vox.com/future-perfect/2022/12/31/23509593/2022-predictions-macron-bolsanaro-covid-inflation-unemployment-climate-change-olympics#:~:text=AI%20will%20discover%20a%20new%20drug%20%E2%80%94%20or%20an%20old%20drug%20fit%20for%20new%20purposes%20%E2%80%94%20that%E2%80%99s%20promising%20enough%20for%20clinical%20trials%20(85%20percent)%20%E2%80%94%20RIGHT

https://www.forbes.com/sites/calumchace/2022/02/25/first-wholly-ai-developed-drug-enters-phase-1-trials/

There's already been two discovered in 2021 according to Vox Future Perfect, and given DeepMind is entering the area, I think the market is underpricing this.

I'm a bit confused by this, because I could sworn a few machine learning methods had already uncovered some candidates. Does it have to be AI designed, not an ML improvement to docking search methods?