Google’s Adam Elman on how AI can help to accelerate the energy transition

Google’s head of sustainability for the EMEA region, Adam Elman, speaks exclusively with edie about the ways in which artificial intelligence (AI) and other digital technologies can help to accelerate the energy transition – and why big tech firms must ensure they aren’t used to fuel the climate crisis.

Google’s Adam Elman on how AI can help to accelerate the energy transition

The International Energy Agency (IEA) is forecasting that renewables will account for 90% of global capacity additions in the electricity sector from now to 2027, with solar being the biggest growth area. Combined with nuclear, this means that renewables will account for more than 90% of the growth in electricity demand, by IEA estimates.

Critics of renewable energy will often point to the fact that the wind does not always blow and the sun does not always shine. There are several comebacks which sustainability leaders can counter with, including the massive growth in the energy storage pipeline, including the emergence of technologies with less or lithium. There is also the fact that this argument almost never seems to account for the need to bring gas and nuclear facilities offline for maintenance periodically, meaning that no facility can guarantee 24/7 energy generation for its whole working life.

Soon, though, there will be another comeback – that AI has helped us to get better at predicting when the wind will blow and the sun will shine, and how we should manage variable generation.

Google’s head of sustainability for the EMEA region, Adam Elman, reflects on a trial that the business has been undertaking with Engie since last summer, conducted to assess the ways in which AI can optimize wind farms.

Google Cloud’s AI Services and Industry Solutions team has developed a bespoke technology solution that enables Engie to predict how much electricity generated using wind turbines should be sold on which power markets, at what time, and at what price point.

Engie is, at present, still trialling the solution on a small portfolio of turbines. But Elman said his team “thinks technology like this certainly has the opportunity to scale”. Google is by no means the only company innovating in this space. Precedence Research stated last year that the global market for AI in renewables is set to grow by more than one-quarter each year through to 2030, reaching almost $76bn by the end of the decade.

Maximising positive impact

Elman, speaking ahead of his appearance at edie 23 this week (scroll down for details), explains that major components of Google’s global sustainability are “supporting partners” to decarbonise and “enabling everyone” to make choices that lower emisisons.

It has made commitments to reduce its own direct footprint, described internally as the “leading at Google” part of the sustainability strategy. A commitment is in place to at least halve emissions across all scopes by 2030, against a 2019 baseline, before insetting and offsetting to address residual emissions and reach net-zero.

But the impact this will have pales in to comparison with the potential impact Google could have by accelerating decarbonisation at the “many thousands of stakeholders” it works with, “from governments and cities, to big companies, startups and not-for-profits” – Engie being just one example.

The impact can be amplified further through the development of tools, both free and paid-for, that either enable people and organisations to make lower-emission choices or select this choice for them. If you use Google Maps, for example, you may have noticed the addition of route planning based on journey emissions and fuel efficiency.

With this level of impact on the daily running of organisations and on daily lifestyles, the opportunities for tech majors like Google to have a cascading positive impact on the energy transition is sizeable. But there are also chances that this scale and reach could be used for the wrong reasons – to place short-term profits before all else, regardless of the impacts of climate, nature and society.

AI has been in the news lately with debates around whether AI like ChatGPT will put workers such as journalists out of work and present new challenges in universities and schools, like plagiarism risks and reduced critical thinking. There has also been a backlash against the use of AI-enabled art creation filters on social media apps, with questions raised about what their creators do with face data and whether the art used in the first instance was ascertained with permission of the artist.

In the sustainability space, there are fears that, in as much as AI can make renewable energy generation and use more efficient, is also being used by energy majors for fossil fuel exploration and extraction. It could be used to reforest, or to deforest.

“This is a hugely important topic, and one that we are extremely aware of – and have been for some time,” Elman tells edie. “We think that AI has potentially tremendous innovation opportunity, particularly in the sustainability and climate space.

“We also recognise that AI and other advanced technologies raise important challenges that we need to address clearly, thoughtfully, and affirmatively.

“One of our key principles is that AI should be socially beneficial, which means that, as we consider developments in and potential uses of AI technologies, we account for a broad range of social and economic factors before we proceed.”

Climate-related factors are on the agenda here, and there are some cases where they are a hard line. Google made a new set of commitments on the responsible use of digital technologies in spring 2020, including a commitment to end the development of bespoke AI solutions for the fossil fuel sector that could be used to maximise oil and gas extraction or aid exploration.

edie asks Elman whether Google and other tech majors can explore partnerships to implement similar commitments on the environmentally responsible use of digital technology. He says this is “entirely the case”, although his employer has not yet launched or joined an initiative on the matter publicly.

Clean energy traceability

So, AI could help to make clean energy systems as efficient and profitable as possible. But other digital technologies also present other opportunities to solve challenges relating to renewable energy uptake, including the traceability of energy and, in turn, the credibility of renewable energy claims.

Most renewable energy tariffs offered to businesses and homes do not provide them directly with renewables. Instead, they are provided with electricity from the grid, which will likely be from a multitude of sources, and their energy provider purchases certificates for renewable electricity generation matching the level of customer consumption.

There is ample opportunity for miscalculation in this system, whether on purpose or due to complexities. Even when the numbers add up, they could relate to existing arrays that do nothing to enable the energy transition in the region – or even the nation – in which the energy user is based. The UK Government notably launched a review of green electricity tariffs and how they are marketed almost two years ago, amid concerns about the use of more renewable energy guarantees of origin (REGO) certificates than should be possible.

Google has been matching 100% of its annual electricity use with renewable energy purchases globally since 2017. But, by 2030, it wants all of its energy demands to be met with renewable energy tracked and verified as locally and recently generated.

Elman explains: “Whilst we are really proud of our work so far… we recognise that this still means that we are looking – at the end of the year – at how much energy we use and whether we’ve purchased enough renewable electricity to match that.

“In some locations, at some times of the day, renewable energy is not being produced, so we’re having to rely on what’s on the grid. We don’t think that’s good enough. We really want to make sure we are sourcing clean energy locally, every hour of the day.”

To achieve this goal, Google has developed a tool that enables itself and other businesses – plus other organisations including governments and energy system operators – to track renewable energy generation and consumption in real-time, in 30 to 60-minute intervals. The tool, called ‘Time-Based Energy Attribute Certificates’ (T-EACs), was first piloted in 2021 and large-scale trials within and beyond Google are now underway.

Elman highlights Google’s wish to help other organisations make the same journey. He says: “In 2030, if we achieve our goal on 24/7 carbon-free energy, but no one else has, then we’ve failed. We really do need system change.”

Last October, Google announced a partnership with C40 Cities on a programme to enable cities to reach  24/7 clean energy generation. London, Paris and Copenhagen are the first three cities taking part, but C40 has almost 100 cities as members, meaning the potential for a rollout is significant.

Google has already worked with the UN and Sustainable Energy for All to create a 24/7 Carbon-Free Energy Compact. The Compact has published a set of principles and actions that stakeholders across energy systems can commit to, and Elman notes that this is “getting real traction”. Compact members cover all parts of the energy system, from governments and system operators, to private generation companies and commercial energy users. To name but a few, the Governments of Iceland and Scotland are participating, as are Microsoft, SSE and Orsted.

As well as providing the framework of principles and actions, the Compact hosts a free academy. Open to members in Europe in the first instance, the hub provides information to answer all frequently asked questions for energy suppliers and buyers. The aims of the hub are awareness raising, supporting the development of schemes and advocating for stronger policy support.

Hear more from Google’s Adam Elman at edie 23 this week

Taking place in London on 1-2 March 2023, edie’s biggest annual event has undergone a major revamp to become edie 23, with a new name, new venue, multiple new content streams and myriad innovative event features and networking opportunities.

edie 23 will take place at the state-of-the-art 133 Houndsditch conference venue in central London. Held over two floors, the event will offer up two full days of keynotes, panels, best-practice case studies and audience-led discussions across three themed stages – Strategy, Net-Zero and Action.

Click here for full information and to book your ticket.

Google’s Adam Elman will be speaking on Day Two of edie 23 (2 March), delivering a case study on how AI can accelerate corporate climate action in the Action Theatre. Springwise’s chairman James Bidwell will be chairing this session.

Action inspires action. Stay ahead of the curve with sustainability and energy newsletters from edie