logo
Your AI prompts could have a hidden environmental cost

Your AI prompts could have a hidden environmental cost

CTV News23-06-2025
Whether it's answering work emails or drafting wedding vows, generative artificial intelligence tools have become a trusty copilot in many people's lives. But a growing body of research shows that for every problem AI solves, hidden environmental costs are racking up.
Each word in an AI prompt is broken down into clusters of numbers called 'token IDs' and sent to massive data centers — some larger than football fields — powered by coal or natural gas plants. There, stacks of large computers generate responses through dozens of rapid calculations.
The whole process can take up to 10 times more energy to complete than a regular Google search, according to a frequently cited estimation by the Electric Power Research Institute.
So, for each prompt you give AI, what's the damage? To find out, researchers in Germany tested 14 large language model (LLM) AI systems by asking them both free-response and multiple-choice questions. Complex questions produced up to six times more carbon dioxide emissions than questions with concise answers.
In addition, 'smarter' LLMs with more reasoning abilities produced up to 50 times more carbon emissions than simpler systems to answer the same question, the study reported.
'This shows us the tradeoff between energy consumption and the accuracy of model performance,' said Maximilian Dauner, a doctoral student at Hochschule München University of Applied Sciences and first author of the Frontiers in Communication study published Wednesday.
Typically, these smarter, more energy intensive LLMs have tens of billions more parameters — the biases used for processing token IDs — than smaller, more concise models.
'You can think of it like a neural network in the brain. The more neuron connections, the more thinking you can do to answer a question,' Dauner said.
What you can do to reduce your carbon footprint
Complex questions require more energy in part because of the lengthy explanations many AI models are trained to provide, Dauner said. If you ask an AI chatbot to solve an algebra question for you, it may take you through the steps it took to find the answer, he said.
'AI expends a lot of energy being polite, especially if the user is polite, saying 'please' and 'thank you,'' Dauner explained. 'But this just makes their responses even longer, expending more energy to generate each word.'
For this reason, Dauner suggests users be more straightforward when communicating with AI models. Specify the length of the answer you want and limit it to one or two sentences, or say you don't need an explanation at all.
Most important, Dauner's study highlights that not all AI models are created equally, said Sasha Luccioni, the climate lead at AI company Hugging Face, in an email. Users looking to reduce their carbon footprint can be more intentional about which model they chose for which task.
'Task-specific models are often much smaller and more efficient, and just as good at any context-specific task,' Luccioni explained.
If you are a software engineer who solves complex coding problems every day, an AI model suited for coding may be necessary. But for the average high school student who wants help with homework, relying on powerful AI tools is like using a nuclear-powered digital calculator.
Even within the same AI company, different model offerings can vary in their reasoning power, so research what capabilities best suit your needs, Dauner said.
When possible, Luccioni recommends going back to basic sources — online encyclopedias and phone calculators — to accomplish simple tasks.
Why it's hard to measure AI's environmental impact
Putting a number on the environmental impact of AI has proved challenging.
The study noted that energy consumption can vary based on the user's proximity to local energy grids and the hardware used to run AI models.That's partly why the researchers chose to represent carbon emissions within a range, Dauner said.
Furthermore, many AI companies don't share information about their energy consumption — or details like server size or optimization techniques that could help researchers estimate energy consumption, said Shaolei Ren, an associate professor of electrical and computer engineering at the University of California, Riverside who studies AI's water consumption.
'You can't really say AI consumes this much energy or water on average — that's just not meaningful. We need to look at each individual model and then (examine what it uses) for each task,' Ren said.
One way AI companies could be more transparent is by disclosing the amount of carbon emissions associated with each prompt, Dauner suggested.
'Generally, if people were more informed about the average (environmental) cost of generating a response, people would maybe start thinking, 'Is it really necessary to turn myself into an action figure just because I'm bored?' Or 'do I have to tell ChatGPT jokes because I have nothing to do?'' Dauner said.
Additionally, as more companies push to add generative AI tools to their systems, people may not have much choice how or when they use the technology, Luccioni said.
'We don't need generative AI in web search. Nobody asked for AI chatbots in (messaging apps) or on social media,' Luccioni said. 'This race to stuff them into every single existing technology is truly infuriating, since it comes with real consequences to our planet.'
With less available information about AI's resource usage, consumers have less choice, Ren said, adding that regulatory pressures for more transparency are unlikely to the United States anytime soon. Instead, the best hope for more energy-efficient AI may lie in the cost efficacy of using less energy.
'Overall, I'm still positive about (the future). There are many software engineers working hard to improve resource efficiency,' Ren said. 'Other industries consume a lot of energy too, but it's not a reason to suggest AI's environmental impact is not a problem. We should definitely pay attention.'
Orange background

Try Our AI Features

Explore what Daily8 AI can do for you:

Comments

No comments yet...

Related Articles

Should You Buy Nvidia Before Aug. 27?
Should You Buy Nvidia Before Aug. 27?

Globe and Mail

time26 minutes ago

  • Globe and Mail

Should You Buy Nvidia Before Aug. 27?

Key Points Nvidia reached a record high recently, and its market value soared to $4 trillion. The company also has reported positive news recently, such as its upcoming return to China's AI chip market. 10 stocks we like better than Nvidia › Nvidia (NASDAQ: NVDA) has achieved many milestones in recent quarters: The top artificial intelligence (AI) chip designer was invited into the Dow Jones Industrial Average last year and ended up delivering the best annual performance in the index. This year, Nvidia slipped ahead of Microsoft to become the world's biggest company, and just days ago it took this a step further by becoming the first company to ever reach $4 trillion in market value. That's as the stock hit a new record high. The company has been on fire for most of the past two years, and right now you may be wondering whether you should hop on board and if so, when. Well, Nvidia, has yet another potential catalyst ahead on Aug. 27. Should you buy the stock before then? Let's find out. Where to invest $1,000 right now? Our analyst team just revealed what they believe are the 10 best stocks to buy right now. Learn More » Seeing the AI potential early Let's start off by exploring Nvidia's story so far. The company has actually been around for quite a long time -- more than 30 years -- but it saw growth surge in recent years as part of the AI boom. Nvidia, seeing the potential of AI, decided early on to design its graphics processing units (GPUs) specifically to meet the needs of this growing industry. And this proved to be a game-changing move for the company. Thanks to this early and strong entrance into AI, Nvidia has seen revenue soar in the double and triple digits, and it's built a rock-solid reputation in the field. Customers needing the most powerful and efficient chips to drive their projects automatically turn to Nvidia, and that's resulted in demand that's even exceeded supply. This helped Nvidia grow annual revenue from about $27 billion two years ago to more than $130 billion in the most recent fiscal year. Net income has climbed too, reaching record levels as Nvidia ensures high profitability on sales of its chips and related products. Nvidia also has made another smart move in recent years, and that's the expansion into a wide variety of AI products and services, making it the go-to seller of anything a customer may need. This includes tools needed for the AI of the future too, meaning Nvidia's revenue power should continue for years to come. For example, Nvidia -- as it powers so many AI functions -- is well positioned to play a key role as agentic AI takes off and if humanoid robots become a growth area down the road. What's happening on Aug. 27? So, now, let's consider what's set to happen on Aug. 27, and that's the tech giant's second-quarter earnings report. A couple of months ago, I wrote about Nvidia stock's performance after earnings reports, and it's shown a trend of gains. That may offer investors a reason to be optimistic about post-earnings performance. It's also important to note that Nvidia received important good news recently. The company, after temporarily losing access to the Chinese market due to U.S. export controls, will be able to sell its GPUs to China soon. The U.S. assured Nvidia that it would receive the license needed to do so. Any details on progress in that area could lift the stock as China is a big market for the company, representing 13% of revenue last year. Nvidia chief Jensen Huang may offer investors a glimpse into the rollout of Blackwell Ultra, the company's latest chip. CoreWeave, a cloud provider of compute for AI, recently said it's now making Ultra available to customers. Huang is known for delivering earnings calls filled with news and elements that offer visibility on what's to come, so comments from him could impact stock performance. Of course, Nvidia stock already has climbed in the double digits this year, and potential general economic or import tariff headwinds could represent risks to performance. So, it's never a good idea to buy a stock now and count on gains in the short term. Instead, it's a better idea to consider that player's long-term potential. This greatly increases your chances of scoring a win. You are likely to post a bigger win than if you buy and sell over a period of a couple of months. All of this means that, though I consider Nvidia a buy right now, you don't have to rush to get in on the stock before the company's earnings report. If you invest ahead of that moment or afterward, the stock could add significantly to the value of your portfolio over the long run. Should you invest $1,000 in Nvidia right now? Before you buy stock in Nvidia, consider this: The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Nvidia wasn't one of them. The 10 stocks that made the cut could produce monster returns in the coming years. Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you'd have $652,133!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you'd have $1,056,790!* Now, it's worth noting Stock Advisor's total average return is 1,048% — a market-crushing outperformance compared to 180% for the S&P 500. Don't miss out on the latest top 10 list, available when you join Stock Advisor. See the 10 stocks » *Stock Advisor returns as of July 21, 2025 Adria Cimino has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Microsoft and Nvidia. The Motley Fool recommends the following options: long January 2026 $395 calls on Microsoft and short January 2026 $405 calls on Microsoft. The Motley Fool has a disclosure policy.

Prediction: These 3 Market-Leading Artificial Intelligence (AI) Stocks Will Eventually Plunge 50% (or More)
Prediction: These 3 Market-Leading Artificial Intelligence (AI) Stocks Will Eventually Plunge 50% (or More)

Globe and Mail

timean hour ago

  • Globe and Mail

Prediction: These 3 Market-Leading Artificial Intelligence (AI) Stocks Will Eventually Plunge 50% (or More)

Key Points Artificial intelligence (AI) is expected to positively impact corporate America's growth arc, with one estimate calling for a $15.7 trillion boost to global gross domestic product by 2030. Despite competitive and/or first-mover advantages, the industry-leading companies on this list are sporting historically unsustainable valuation multiples. Additionally, history suggests an AI bubble will eventually form and burst, which would be problematic for all three of these market-leading businesses. 10 stocks we like better than Palantir Technologies › Not since the advent of the internet has a technological revolution come along that promises to bump up the growth arc for corporate America. The analysts at PwC believe the combination of consumption-side effects and productivity improvements traced specifically to the rise of artificial intelligence (AI) will add $15.7 trillion to global gross domestic product in 2030. With an addressable market this massive, it's no surprise to see public companies involved in AI hardware and applications rocketing higher. However, next-big-thing trends have historically had a knack for giving way to next-big-thing bubbles -- and artificial intelligence is unlikely to be the exception to this unwritten rule. Some of today's biggest AI leaders may be poised to become tomorrow's losers. What follows are three market-leading AI stocks with mounting headwinds that can eventually (i.e., I'm not trying to time their respective peaks) plunge 50% or more. Palantir Technologies The first megacap AI stock that can plummet at least 50% in the coming months or years is data-mining specialist Palantir Technologies (NASDAQ: PLTR). Whereas some companies I'm, admittedly, not a fan of, Palantir is what I'd consider an excellent business with a completely unsupported valuation. What makes Palantir a great company is its AI- and machine learning-driven software-as-a-service platforms, Gotham and Foundry. The former is used by federal governments for mission planning and execution, along with data gathering and analytics. As for Foundry, it makes sense of big data for businesses. Both of these segments are growing by sustained double-digit percentages and neither are replicable at scale by other companies. While companies with sustainable competitive advantages are often bestowed with a valuation premium by Wall Street, relative to their peers, Palantir's premium is otherworldly. Looking back more than three decades at other megacap companies whose bubbles burst while on the leading edge of a next-big-thing trend shows price-to-sales (P/S) multiples typically peak between 30 and 40. Palantir's P/S ratio closed out July 18 at almost 123. Even if Palantir were to grow its sales by 30% annually over the next five years and its stock went sideways, it would still be in bubble territory. The other significant worry for Palantir is Gotham. Though this segment is primarily responsible for lifting Palantir to recurring profitability, there's a very limited pool of federal governments that can access this platform. Further, it's unclear if government spending on defense will remain robust beyond President Donald Trump's second term. Palantir is priced for perfection amid a growth trend that's anything but perfect. Tesla Whereas I view Palantir Technologies as a phenomenal business with an unsightly valuation, I struggle to find many redeeming qualities for electric-vehicle (EV) maker Tesla (NASDAQ: TSLA), whose EVs and full self-driving (FSD) software rely on AI and technology to safely navigate obstacles. Tesla became one of Wall Street's most-influential businesses by riding its first-mover competitive advantages in the EV space. Under the leadership and vision of CEO Elon Musk, it expanded production and deliveries to approximately 1.8 million EVs in each of the last two years. More importantly, Tesla has been profitable on a recurring basis for five consecutive years, with Musk overseeing the expansion of its energy generation and storage operations. This segment has the potential to minimize the cyclical ebbs-and-flows associated with Tesla's auto industry operations. The issues I have with Tesla are threefold. To begin with, EV growth has completely stalled. The selling price of its four primary models (3, S, X, and Y) were slashed on more than a half-dozen occasions due to increasing EV competition and rising inventory levels. This has wreaked havoc on Tesla's vehicle margin. Secondly, Tesla's earnings quality is poor. Though the company remains decisively profitable, more than half of its pre-tax income has consistently come from automotive regulatory credits given to it for free by governments and net interest income earned on its cash. President Donald Trump's One Big, Beautiful Bill removes these U.S. regulatory credits, and the Federal Reserve is currently in a rate-easing cycle, which threatens to lower future interest income-earning potential. In other words, non-innovative channels are driving Tesla's profits. Lastly, I view Musk as more of a lability than an asset to Tesla. His habit of overpromising and underdelivering on innovations has grown stale. Tesla's robotaxi launch was underwhelming; its FSD software hasn't reached Level 5 autonomy as promised more than a decade ago; and Cybertruck sales have come in well below expectations. If these unfulfilled promises are backed out of Tesla's market cap, its stock would fall far more than 50%. Nvidia I know it's Wall Street blasphemy to say this, but the third AI stock that can eventually plunge 50% or more is the face of this technological revolution, Nvidia (NASDAQ: NVDA). Nvidia gets lumped in with Palantir in the sense that I recognize it's a phenomenal business, but its valuation (when combined with other headwinds) makes absolutely no sense. Nvidia's claim to fame is that its Hopper and Blackwell graphics processing units (GPUs) are the clear top options by businesses operating AI-accelerated data centers. With supply chains unable to keep pace with overwhelming demand for AI hardware, Nvidia has been able to sell more of its chips each quarter, as well as charge a triple-digit premium for its hardware, relative to its direct external competitors. Additionally, Nvidia should be able to sustain its compute advantages. CEO Jensen Huang expects to bring a new advanced AI chip to market annually, with Blackwell Ultra (2025), Vera Rubin (2026), and Vera Rubin Ultra (2027) following Hopper and Blackwell. It's unlikely that any external competitors come close to the compute capabilities of Nvidia's hardware. But as I pointed out earlier, there hasn't been a next-big-thing technology in decades that's avoided an early stage bubble-bursting event. Investors consistently overshoot when it comes to early adoption rates and utility for new innovations -- and AI is unlikely to be the exception. If an AI bubble forms and bursts, arguably no stock would be hit harder than Nvidia. Competition is another clear issue for Nvidia. Despite maintaining its compute advantages, Nvidia can still lose out on valuable data center real estate. The biggest threat just might be that many of its largest customers by net sales are internally developing AI-GPUs to use in their data centers. These chips are less costly and more readily available than Nvidia's often-backlogged hardware. Rounding things out, Nvidia is nearing a P/S ratio of almost 29, which borders on the aforementioned bubble territory I discussed earlier. Should you invest $1,000 in Palantir Technologies right now? Before you buy stock in Palantir Technologies, consider this: The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Palantir Technologies wasn't one of them. The 10 stocks that made the cut could produce monster returns in the coming years. Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you'd have $652,133!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you'd have $1,056,790!* Now, it's worth noting Stock Advisor's total average return is 1,048% — a market-crushing outperformance compared to 180% for the S&P 500. Don't miss out on the latest top 10 list, available when you join Stock Advisor. See the 10 stocks » *Stock Advisor returns as of July 21, 2025

Cisco Takes Licence to the Sisvel Wi-Fi 6 Patent Pool
Cisco Takes Licence to the Sisvel Wi-Fi 6 Patent Pool

National Post

timean hour ago

  • National Post

Cisco Takes Licence to the Sisvel Wi-Fi 6 Patent Pool

Article content LUXEMBOURG — Cisco Systems, Inc has become the latest licensee of the Sisvel Wi-Fi 6 patent pool programme. Article content The deal, which was reached on an amicable basis, provides Cisco with one-stop access to close to 2,000 patents (245 patent families) recognised to date, through the pool's process of independent patent evaluations, as essential to the 802.11ax Wi-Fi 6 standard. Article content Cisco is the worldwide technology leader that is revolutionising the way organisations connect and protect in the AI era. It is a global technology leader in enterprise networking, including enterprise grade access points. Article content The Sisvel Wi-Fi 6 Pool was established in July 2022 and has seen a significant surge in interest recently. In addition to Cisco, more than 20 new licensees have come on board since the start of Q4 2023. Article content The patents in the Sisvel Wi-Fi 6 pool are held by the companies identified at All of them have played key roles in the development of Wi-Fi 6 technology. Article content 'We are delighted to welcome Cisco as a licensee of our Wi-Fi pool and thank them for the professional way in which they have handled the negotiation process,' says Sisvel's Head of Licensing, Nick Webb. 'This deal is a significant milestone for the pool and a major validation of its proposition, which is to offer transparent and efficient access to patents that underpin world class Wi-Fi technology. We look forward to welcoming further licensees over the coming months and to powering Wi-Fi innovation for many years to come.' Article content Article content Article content Article content Article content Article content Article content

DOWNLOAD THE APP

Get Started Now: Download the App

Ready to dive into a world of global content with local flavor? Download Daily8 app today from your preferred app store and start exploring.
app-storeplay-store