Post here all non technical related topics about Formula One. This includes race results, discussions, testing analysis etc. TV coverage and other personal questions should be in Off topic chat.
So some answers about the use of 'AI' (or let's say ML/DL) for aero development :
At least 6 F1 teams are using it (probably all of them by now, maybe with a different supplier) Biggest advantages are : instant result as opposed to hours or days, and considered as post-processing so no limits on number of predictions as opposed to CFD / wind-tunnel runs Also, makes collaboration between engineering teams (aero, design, structure) much more easy. Instant results means no need to wait for the next meeting, waiting for the results Some explanations on inputs : parameters vs geometries. ML vs DL Training the model. Quality and distribution VS quantity. Teams tend to use the data they already have available and deal with it despite the greater uncertainty Dealing with uncertainties with a Traffic Light System Other uses : crash test, acoustics, probably strategies etc. Just one more tool for the engineers, still people to take the decision after the predictions are done
AI has been used already to design lighter parts, stressed parts, for racing engines
AI was used to build a complete sports car
I think the main issue with using AI is that they will completely deplete the processing power allocation on training and validating any relevant model, if the aim is to design the whole thing.
On smaller items or parts...might be possible!
It doesn't have to give the final iteration to be useful
Can try 1000 and answer with the best 5 or 10 to be run on CFD
At least 6 F1 teams are using it (probably all of them by now, maybe with a different supplier)