Corporations offer academics vast proprietary data, tools and resources that can accelerate research and innovation, but these partnerships generate tensions over competing interests, openness, and ethical concerns. Major technology firms and AI companies now spend R&D budgets that far exceed most academic labs, shifting investments toward research. High-profile advances such as AlphaFold demonstrate research benefits from industry resources. Successful collaborations require acknowledging potential conflicts and implementing formal procedures for resolving them, as seen in organized engineering-research hubs, to balance corporate goals, academic integrity, transparency and societal ethical considerations such as misinformation, bias and defence affiliations.
Over the past few decades, I have seen how challenging such partnerships can be. But I've also seen how following best practices can make them work. In 2014, I co-authored Organized Innovation, analysing the US Engineering Research Centers programme - supporting hubs hosted in universities and funded by the US National Science Foundation that engage productively with corporations. We learnt that successful collaborations not only acknowledge that conflicts can arise, but also commit to creating formal procedures for resolving them.
The stakes are high today. The 'magnificent seven' technology firms - Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla - and newcomers such as OpenAI have research and development (R&D) budgets that dwarf spending available to most academic laboratories. In 2022, Amazon, Apple, Google, IBM and Microsoft spent tens of billions of dollars on R&D. In 2024, Meta alone spent nearly US$44 billion, or 27% of its annual revenues, on R&D.
Many tech investments have shifted from the development to the research side. Last year, Demis Hassabis, co-founder of the artificial intelligence (AI) start-up firm DeepMind, acquired by Google in 2014, received jointly with his colleague John Jumper half of the Nobel Prize in Chemistry for developing AlphaFold, an AI model that predicts protein structures. Yet, many academics remain wary of working with 'big tech', whose approaches can be controversial.
Collection
[
|
...
]