Nscale’s Post

There is a lot of conversation at the moment about the environmental impact and energy consumption of AI. “The reality is AI workloads are so often poorly optimised, wasting energy use and delivering poor returns. Accelerating AI #hardware and #software optimisation is one of the best routes to managing AI energy consumption levels” - Elio VP At Nscale, our vertically integrated platform ensures every layer of the AI stack is optimised for efficiency and performance, helping organisations achieve their #AI and decarbonisation goals. Find out more: https://lnkd.in/edJ6Tubg #Sustainability #Technology #Innovation

#Microsoft announced a 30% increase in emissions since 2020, primarily due to the expanded infrastructure that AI and cloud computing systems run on. The jump in emissions caused by AI, is yet to match the hype of AI itself right now. We will soon expect increased scrutiny from consumers, governments and civil society. If Microsoft experiences this, it is fair to assume similar trends for many other organizations worldwide that are also incorporating AI in their strategy. Society needs to find a more responsible approach to utilize AI sustainably. The first (easier) step is to publish the data and acknowledge its severity, which Microsoft rightfully did. The next (harder) step is to take actions to change the status quo. Microsoft aims to require certain suppliers to use 100% carbon-free electricity by 2030 - a bold yet critical claim that will be closely watched. On the receiving end of AI-enabled services, what can consumers do? How will governments regulate these developments? CNBC: https://lnkd.in/dR8Efg_e FT: https://lnkd.in/dC_ZCUPK

The Double-Edged Sword: Google's AI Push and the Race to Net-Zero Google's ambitious goal of achieving net-zero emissions by 2030 faces a formidable challenge: the ever-growing power demands of artificial intelligence (AI). A recent report revealed a 48% rise in Google's greenhouse gas emissions since 2019, with data centers identified as the main culprit. AI's thirst for processing power translates to a surge in energy consumption, particularly in data centers. This is a looming concern not just for Google, but for the entire tech industry. A study by the International Energy Agency predicts a potential doubling of global data center electricity use by 2026. This isn't slowing Google down though. In January, they committed a staggering $1 billion to build a new data center in the UK, further amplifying their AI capabilities. The report acknowledges the tightrope Google walks. While emissions rose 13% year-over-year in 2023, Google emphasizes its strides towards clean energy. Data centers in Europe and the Americas are increasingly powered by carbon-free sources. However, the picture is far from uniform. Regions like Asia-Pacific lag behind in clean energy availability, creating significant roadblocks. Despite the challenges, Google remains committed. They recognize the need to bridge the gap between clean energy investment and actual emissions reduction. This transparency, alongside their continued efforts, is a ray of hope in the race for a sustainable future.

Artificial Intelligence puts Google and Microsoft's emission 'moonshot' targets on a different planet. 🚀 Google just reported a 13 percent rise in greenhouse gas emissions for 2023. This not long after Microsoft's report mid-May where it posted a 29 percent increase in its emissions since 2019. Where Google's emissions grew 48% over those same 5 years. Rising emissions driven by the energy appetite of #artificialintelligence and scarce availability of renewable energy in Asia and certain U.S. regions. Both companies are struggling to add renewable energy fast enough to meet the explosive appetite for AI. It highlights their challenge of reducing emissions while compute intensity increases and the need to grow their technical infrastructure investments in data centers to support this global AI transition. Google is betting big on AI for its cloud services business and as the foundation for future search revenue. It is also a big focus for their sustainability team, because of two reasons: 🌍 AI for Sustainability. Google puts AI to work to help cities, companies and individuals reduce their carbon footprints by 1 gigaton by 2030. The company cites research estimating that AI could help mitigate 5-10 percent of global emissions by the end of the decade. 📈 Manage AI's environmental impact. It prioritizes model optimization to reduce AI's "training" time and resources, more efficient infrastructure, and emission reductions through investments in renewable energy. Google already stated that their goal of reaching net zero emissions by 2030 "won't be easy". Where Microsoft's president stated about their "moonshot" target that "the moon has moved" due its AI strategy. Do we remain hopeful about AI's potential to drive positive change? Are we vigilant and clear-eyed enough about its potential environmental impact? What do you think? :)

Google wanted to be net zero for carbon in 2030. Google's carbon usage has gone up 48% since 2019. AI searches are estimated to taken ten times the energy of a standard search. The logic displayed in attempting to justify is that AI will save energy. Just feels like boiling frogs discussing hot tub design https://lnkd.in/eCx48_J3

Hey everyone, just came across this interesting article about Google's greenhouse gas emissions shooting up by 48% in the past 5 years. The rise has been attributed to the company's rapid expansion in cloud computing and AI. It's a stark reminder of the environmental impact of technology and the urgent need for sustainable solutions. Let's continue to stay informed and push for greener practices in tech. #GreenTech #ClimateAction #SustainabilityAwareness https://ift.tt/98nIwtR

Google's AI Focus and Sustainability Goals: A Balancing Act A recent report by Google highlights a growing challenge: the company's significant investments in artificial intelligence (AI) are impacting its progress towards sustainability goals. While Google has set an ambitious target of reducing greenhouse gas emissions by 50% by 2030 (compared to 2019 levels), their environmental report reveals a 48% increase in emissions since the baseline year. Data centres, particularly those supporting AI development, are identified as a key contributor to this rise. The report details a 17% increase in data centre energy consumption in 2023, with this trend expected to continue. This issue extends beyond Google, with the International Energy Agency (IEA) predicting a potential tenfold increase in the AI industry's energy demands by 2026. Such a surge raises concerns about grid stability and a potential reliance on fossil fuels for power generation. Commendably, Google acknowledges this challenge and outlines efforts to improve energy efficiency within its AI models and data centres. Additionally, they remain committed to achieving carbon-neutral operations by 2030 through the use of renewable energy sources. However, a critical question remains: can the company effectively balance its pursuit of AI innovation with its environmental sustainability goals? This complex issue invites industry-wide discussion. #climatechange #artificialintelligence #sustainability #Google

𝐓𝐡𝐞 𝐝𝐢𝐫𝐭𝐲, 𝐧𝐨𝐭-𝐬𝐨-𝐬𝐞𝐜𝐫𝐞𝐭, 𝐭𝐫𝐮𝐭𝐡 𝐚𝐛𝐨𝐮𝐭 𝐀𝐈'𝐬 𝐢𝐦𝐩𝐚𝐜𝐭 𝐨𝐧 𝐭𝐡𝐞 𝐞𝐧𝐯𝐢𝐫𝐨𝐧𝐦𝐞𝐧𝐭 In this important linked article, Camilla Hodgson of the Financial Times writes: "Microsoft’s emissions have risen by almost a third since 2020, as the push to build out the infrastructure behind artificial intelligence threatens its climate goals. The nearly 30 per cent increase in emissions was in large part due to the construction of the data centres that AI and cloud computing systems run on, Microsoft said in its annual sustainability report on Wednesday. " The links from #artificialintelligence, particularly #generativeAI, to AI #compute requirements (i.e., the physical infrastructure and computer hardware, together with software and processes that enable AI), to AI #cloudinfrastructure, processing demands and #cloudservices, to the need for ever more and more powerful #datacentres (AI power demands are crowding out other #energy needs), to the the #BigTech race to dominate the #AI future (more, more, more until we win) are not a secret. Another aspect is the whole AI compute and development #supplychain. Hodgson reports that Microsoft's own direct and energy-related #emissions fell by 6.3 per cent in 2023 compared with its 2020 baseline. But emissions from Microsoft's supply chain — which make up the biggest part of its total emissions — jumped by 30.9 per cent, increasing overall emissions to 29.1%. OK, so, as Hodgson writes, Microsoft - like other Big Tech players - has set various #climategoals, including targets to become “carbon negative” and achieve “zero waste” by 2030. But, as she notes: "... those goals have been imperilled by the race to build generative AI, which is power-intensive with large energy and water demands. The competition to build out data centre infrastructure has also raised questions about the capacity of national energy grids to cope with the expected jump in electricity demand linked to AI, and whether there is sufficient renewables generation in those markets to power the technology.". And, as we have seen with #ESG and other causes, in practice #thirdpartysupplychains are actually hard to manage, even for powerful, sophisticated organisations. We're still learning about the currently insatiable demands of AI for more computing power and energy, with the emissions and #pollution that follow. Not everything has yet been revealed. What happens next?

"Google's greenhouse gas emissions in 2023 were 48% higher than in 2019, according to its latest environmental report. The tech giant puts it down to the increasing amounts of energy needed by its data centres, exacerbated by the explosive growth of artificial intelligence (AI)." "AI-powered services involve considerably more computer power - and so electricity - than standard online activity, prompting a series of warnings about the technology's environmental impact. Google's target is to reach net zero emissions by 2030 but it admits that "as we further integrate AI into our products, reducing emissions may be challenging." #ai #energy #electricity #energytransition #energypolicy #infrastructure