REPORTED RATES OF NOVEL MATERIAL DISCOVERY FROM A STUDY OF SCIENTISTS WORKING WITH AND WITHOUT AI TOOLS
$ )/$./. with AI /**'
$ )/$./. 2$#*0/ /**'
INTRODUCITON OF AI TOOL
150
125
100
75
50
25
0
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
MONTHS FROM START OF STUDY
only been battery related. Researchers using AI tools have also found they can engineer enzymes for biofuel synthesis, predict high-yielding biofuel feedstocks, identify industry-beating catalysts for hydrogen-producing electrolysers and generate materials for carbon dioxide (CO2) capture. As AI becomes an increasingly indispensable part of the research process for energy technologies, innovators will also benefit from developments in adjacent areas, including improved robotics and automation. A recent study of the impact of using AI tools in an industrial research setting showed a 39% increase in patenting by the company in under two years. Major obstacles remain Still, serious challenges must be overcome before AI techniques can fulfil their full potential on innovation. One key issue is data availability. Datasets used today have incomplete information
about possible materials and represent a restricted subset of molecules or reactions. The development of massive, structured, specialised datasets to train AI models, such as the Materials Project and Cambridge Structural Database, is underway, but they must be further expanded if real-world scientific problems are to be solved. While creation of “synthetic data” to train models can overcome some of the data gaps, there is no substitute for experimental data, and the fastest route to large and reliable experimental datasets is co-operation between laboratories, including at the international level. The Mission Innovation M4E platform is an example of an international initiative that could demonstrate how governments can support common protocols and jointly curated data. Another challenge is finding ways for AI to optimise results for more than just a narrow set of characteristics and incorporate considerations that are essential for a material to be integrated into a functional product. Today, substantial human checking and testing is still required – for example, to assess performance at different temperatures or interactions with all other components of a device. Also, working out the recipe for manufacturing the materials designed by AI can create considerable follow-on work. Having AI perform these more complex tasks appears feasible, but it leads to high computational requirements and costs that must be assessed.
A recent study of the impact of using AI tools in an industrial research setting showed a 39% increase in patenting by the company in under two years.” “
34
THE FUTURE OF ENERGY
Made with FlippingBook - Online magazine maker