Google quits AI for oil and gas

Anticipating pressure from the environmental movement, Google stops making customized artificial intelligence tools for oil and gas. A Greenpeace report accuses tech firms of ‘helping big oil profit from climate destruction’. The well-documented summary of current oil and gas cloud computing initiatives may make oils regret publicising their computing expertise so widely. ‘Zero Cool’ mole questions Microsoft’s carbon math.

As has been widely reported, Google is to stop making customized artificial intelligence tools for oil and gas firms. Early reports appeared to suggest that Google had caved in to pressure from Greenpeace. In fact, Google’s announcement came before the publication of Greenpeace’s report ‘Tech companies are helping big oil profit from climate destruction ’. According to CNBC, Google Cloud took $65 million from oil and gas companies in 2019, less than 1% of total Cloud revenues. One company that is likely to be impacted by Google’s stance is Total which in 2018 signed with Google to develop artificial intelligence solutions for geoscience.

The Greenpeace report is an extraordinarily well documented summary of current oil and gas cloud computing initiatives which may make oils think twice before publicising their computing expertise. Greenpeace observes that ‘contracts between tech firms and oil and gas companies are now found in every phase of the oil and gas production chain and significantly undermining the climate commitments that Microsoft, Google, and Amazon have made’. Greenpeace cites a popular upstream IT trope that ‘while the near-term outlook for the oil and gas industry is bleak, AI partnerships [… represent …] a critical toolkit that fossil fuel companies will use to bounce back from this downturn’.

It’s not just Google that Greenpeace has in its sights, Amazon and Microsoft are also called-out as ‘partnering with oil companies to use AI to unlock oil and gas deposits in the US and around the world’. The size of the prize for Greenpeace is evidenced from some rather convenient analyses by Accenture ($425 billion from advanced analytics by 2025) and BloombergNEF (annual spend on cloud and advanced analytics at $15.7 billion in 2030).

Greenpeace has it that Microsoft ‘appears to have the most AI contracts with oil and gas companies’. Which means that Microsoft can ‘never truly achieve its recently announced carbon negative’ goal while continuing to aid the oil and gas sector. Amazon chief Jeff Bezos is likewise criticized for allowing AWS to support oil companies while announcing a ‘climate pledge’ and $10 billion ‘earth fund’.

The Greenpeace attack on oil and gas cloud computing is extraordinarily well informed, touching on AI/ML, big data and the internet of things. The study explains in some detail how oils are using cloud technology to plug gaps in data, perform 3D modelling and seismic processing. OSDU, the Open subsurface data universe gets a special mention, as does the fact that all three cloud vendors are on board OSDU.

Greenpeace is on shakier ground when it beats up on the use of high performance computing in leak and spill detection, putting oils (and their cloud service providers) in a ‘damned if you do and damned if you don’t’ position. Greenpeace goes as far as naming some ex-oil company execs who are now heading up cloud initiatives chez the providers. But then Greenpeace is not one for softball tactics – witness the misplaced zeal with which it attacked the Brent Spar decommissioning.

Perhaps the biggest risk for oils lies in Greenpeace’s attack on Microsoft as a major oil and gas service provider. Citing a Microsoft mole called ‘Zero Cool’ and a ‘Microsoft Workers 4 Good’ coalition, Greenpeace thinks it has put its finger on ‘a gaping hole in Microsoft’s carbon math equation’. Greenpeace also calls out Microsoft and Google for rebranding their oil and gas minisites as addressing the ‘energy sector’ (rather than ‘oil and gas’) in a thinly disguised bit of greenwashing.

The report winds-up with a ‘roadmap’ for the cloud providers involving a moratorium on all new machine learning and high-performance computing contracts for oil and gas companies, an environmental impact assessment of ‘the most problematic contracts’, a winding down of existing contracts and a public to ‘pivot’ AI contracts towards the renewable energy sector.

Click here to comment on this article

Click here to view this article in context on a desktop

© Oil IT Journal - all rights reserved.