Could AI put an end to animal testing?

The use of artificial intelligence (AI) in the medical field is revolutionizing the way we approach animal testing. AI systems are now being employed to analyze existing global animal testing results, which helps to eliminate the need for unnecessary new tests. This is particularly beneficial because it can be quite challenging for scientists to sift through years of data to find the information they need. Thomas Hartung, a toxicology professor at Johns Hopkins University, believes that AI is just as good as, if not better than, humans at extracting information from scientific papers. With over 1,000 new compounds entering the market annually, AI systems are now capable of determining the toxicity of these new chemicals. This represents a significant leap forward in terms of both power and accuracy. AI is now involved in every stage of toxicity testing and is even being utilized in the creation of new drugs. However, AI is not without its flaws, particularly when it comes to data bias. For instance, if an AI system has been trained using health data predominantly from one ethnic group, its conclusions may not be suitable for other groups. Despite this, AI is proving to be more accurate than animal testing in many cases. One notable project, AnimalGAN, is aiming to replace the need for future animal testing by accurately determining how rats would react to any given chemical. Another international project, Virtual Second Species, is developing an AI-powered virtual dog to check for potential toxicity in new medicines. While AI testing is making significant progress, obtaining regulatory approval remains a major challenge. There are differing opinions on the future of animal testing, with some advocating for a complete ban, while others argue that it is necessary for many aspects. The hope is that AI can play a role in transitioning away from using animals in tests and experiments.