We Might Have to Accept Some Degree of AI "Not Net Zero"

An argument can be made that artificial intelligence operations will consume vast quantities of electricity and water, as well as create lots of new e-waste. It's hard to argue with that premise. After all, any increase in human activity--including computing intensity--will have that impact.

Some purists might insist we must be carbon neutral or not do AI. Others of us might say we need to make the same sorts of trade offs we must make everyday, for all our activities that have some impact on water, energy consumption or production of e-waste.

We have to balance outcomes and impacts, benefits and costs, while working over time to minimize those impacts. Compromise, in other words.

Some of us would be unwilling to accept "net zero" outcomes if it requires poor people to remain poor; hungry people to remain hungry.

And not all of the increase in e-waste, energy or water consumption is entirely attributable to AI operations. Some portion of the AI-specific investment would have been made in any case to support the growth of demand for cloud computing. 

 So there is a “gross” versus “net” assessment to be made, for data center power, water and e-waste purposes resulting from AI operations. 

By definition, all computing hardware will eventually become “e-waste.” So use of more computing hardware implies more e-waste, no matter whether the use case is “AI” or just “cloud computing.” And we will certainly see more of both. 

Also, “circular economy” measures will certainly be employed to reduce the gross amount of e-waste for all servers. So we face a dynamic problem: more servers, perhaps faster server replacement cycles, more data centers and capacity, offset by circular economy efficiencies and hardware and software improvements. 

Study Name	Date	Publishing Venue	Key Conclusions
The E-waste Challenges of Generative Artificial Intelligence	2023	ResearchGate	Quantifies server requirements and e-waste generation of generative AI (GAI). Finds that GAI will grow rapidly, with potential for 16 million tons of cumulative waste by 2030. Calls for early adoption of circular economy measures.
Circular Economy Could Tackle Big Tech Gen-AI E-Waste	2023	EM360	Introduces a computational framework to quantify and explore ways of managing e-waste generated by large language models (LLMs). Estimates annual e-waste production could increase from 2.6 thousand metric tons in 2023 to 2.5 million metric tons per year by 2030. Suggests circular economy strategies could reduce e-waste generation by 16-86%.
AI has a looming e-waste problem	2023	The Echo	Estimates generative AI technology could produce 1.2-5.0 million tonnes of e-waste by 2030 without changes to regulation. Suggests circular economy practices could reduce this waste by 16-86%.
E-waste from generative artificial intelligence"	2024	Nature Computational Science	Predicts AI could generate 1.2-5.0 million metric tons of e-waste by 2030; suggests circular economy strategies could reduce this by up to 86%1 2
"AI and Compute"	2023	OpenAI (blog)	Discusses exponential growth in computing power used for AI training, implying potential e-waste increase, but doesn't quantify net impact
"The carbon footprint of machine learning training will plateau, then shrink"	2024	MIT Technology Review	Focuses on energy use rather than e-waste, but suggests efficiency improvements may offset some hardware demand growth

Which Firm Will Use AI to Boost Revenue by an Order of Magnitude?

Ultimately, there is really only one way for huge AI infrastructure investments up by an order of magnitude over cloud computing investment to pay off: revenues will have to increase by an order of magnitude as well. 

It might be a stretch to argue that is possible for most firms investing in generative AI frontier models, for example. But it is almost certain that at least one or two of those firms will manage to do so, emerging as leaders of a new industry they lead. 

And that is the prize.  

source: Sherwood 

But generative AI, the current focus of capex, might be--if not a full-blown general-purpose technology--the sort of digital product that creates a whole new--and big--industry. Think of the past pattern of new industries built on firms and products including operating systems, e-commerce, search, social media and online advertising in general, plus still-growing businesses such as ride-hailing and peer-to-peer lodging. 

if generative AI winds up being a “winner take all” business, as most other computing segments have been, there will be no prize for third best, and limited advantage for being second best. 

We have already seen that pattern in many other computing markets. The leader in search has 91 percent market share. The browser leader has 65 percent share. The mobile operating system leader has 72 percent share. The U.S. ride-hailing leader has 68 percent share. 

Market	Dominant Player	Market Share	Runner-up	Market Share
Search Engines	Google	91.9%	Bing	3.0%
Desktop Browsers	Chrome	65.72%	Safari	18.22%
Mobile Browsers	Chrome	66.17%	Safari	23.28%
E-commerce	Amazon	37.8% (US)	Walmart	6.3% (US)
Video Streaming	YouTube	2.5B users	Netflix	231M subscribers
Music Streaming	Spotify	31%	Apple Music	15%
Ride-hailing (US)	Uber	68%	Lyft	32%
Cloud Services	AWS	32%	Azure	22%
Mobile OS	Android	71.8%	iOS	27.6%

So, whether investors like it or not, would-be leaders of the generative AI ecosystem are pouring resources into the effort to lead the new market. And that investment intensity affects investor perceptions, even as the big firms continue to post revenue growth. 

Alphabet reported a robust 15-percent revenue growth; 35-percent cloud computing revenue growth; operating income up 34 percent but also AI-focused capital investment up 72 percent. 

“And as we think into 2025, we do see an increase in AI-focused capital investment coming in 2025,” said Alphabet CFO Anat Ashkenazi. 

So does Amazon, which expects capex to be about  $75 billion in 2024 and “more than that in 2025,” according to Amazon CEO Andy Jassy. “And the majority of it is for AWS and specifically, the increased bumps here are really driven by Generative AI.

“Our AI business is a multi-billion dollar business that's growing triple-digit percentages year-over-year and is growing three times faster at its stage of evolution than AWS did itself,” said Jassy.

Generative AI “is a really unusually large, maybe once-in-a-lifetime type of opportunity,” he said.

All that is fueling investment into generative AI, which based on recent computing product precedent, will produce  a “winner take all” market. 

Company	2024 Estimated AI Capex	2025 Estimated AI Capex
Microsoft	$80 billion	Significant increase
Amazon	$75 billion	Further increase
Alphabet	$52 billion	Increase expected
Meta	$38-40 billion	Significant growth

Microsoft and Meta Platforms both beat analyst expectations with their quarterly earnings reports, but also said more AI spending is coming, pushing down share prices for both firms. 

Microsoft CEO Satya Nadella noted continued capacity constraints at data centers amid surging demand, but also continues heavy spending  on cloud and AI to scale to alleviate capacity constraints. 

Meta CEO Mark Zuckerberg also forecast a "significant acceleration" in spending on AI-related infrastructure in 2025. Zuckerberg acknowledged that this may not be what investors want to hear in the near term, but insisted that the opportunities here "are really big."

GenAI is a big gamble. Based on history, we might suggest that all but one or two of these efforts will fail, and the list of serious contenders also includes OpenAI and others. Should that pattern hold, the top two companies might have 60 percent to 80 percent share of the total market. 

Market Position	Market Share	Profit Share
Leader	70-90%	80-90%
Runner-up	10-20%	5-15%
Others (3-10)	5-10%	0-5%

AI Model Inference Costs Will Decline 20% to 30% Per Year

Despite concern over the high capital investment costs in infrastructure to support generative artificial intelligence models, many studies suggest that costs for inference, which should ultimately be the primary on-going costs, should drop over time, as have costs for other computing instances.

Study Title	Date	Publication Venue	Key Conclusions on Cost Declines
Scaling and Efficiency of Deep Learning Models	2019	NeurIPS	Demonstrates how advances in model scaling (larger models running on optimized hardware) lead to inference cost reductions of around 20-30% per year.
The Hardware Lottery	2020	Communications of the ACM	Highlights the role of specialized hardware (GPUs, TPUs) in reducing AI inference costs, estimating a 2x decrease every 1-2 years with hardware evolution.
Efficient Transformers: A Survey	2021	Journal of Machine Learning Research	Describes optimization techniques (such as pruning, quantization) that contribute to cost declines, estimating an average 30-50% drop in inference costs over two years.
The Financial Impact of Transformer Model Scaling	2022	IEEE Transactions on AI	Examines the economic impacts of scaling transformers and shows that large models can reduce costs by ~40% through efficiencies gained in distributed inference and hardware.
Inference Cost Trends in Large AI Deployments	2023	ICML	Finds a 50% reduction in inference costs per year for large-scale deployments, driven by optimizations in distributed computing and custom AI chips.
Beyond Moore’s Law: AI-Specific Hardware Innovations	2023	MIT Technology Review	Discusses how specialized hardware design reduces inference costs by 2-4x every 2 years, shifting from general-purpose GPUs to domain-specific architectures.
Optimizing Inference Workloads: From Data Center to Edge	2024	ArXiv	Analyzes cost reductions from 2020 to 2024 for both data center and edge deployments, concluding that distributed systems and model compression lead to 50% annual cost drops.

The implication is that inference costs should continue to drop. 

Year	Cost per Inference ($)	Cost Decline Compared to Prior Year
2018	1	-
2019	0.8	20%
2020	0.5	37.50%
2021	0.3	40%
2022	0.15	50%
2023	0.08	47%
2024	0.04	50%
2025	0.02	50%

Of course, a trend towards larger models, using more parameters, will run counter to that trend, in terms of model building. Still, AI model-building (training) cost declines over time, because of hardware acceleration, improved algorithms and model design optimization.

Study	Date	Publication Venue	Key Conclusions on Cost Declines
Scaling Neural Networks with Specialized Hardware	2018	NeurIPS	Describes how hardware advances, especially GPUs and early TPUs, helped reduce model-building costs by around 50% annually for larger models compared to CPU-only setups.
Reducing the Cost of Training Deep Learning Models	2019	IEEE Spectrum	Shows a 40% cost reduction for model training per year through advances in parallel computing and early model optimizations such as batch normalization and weight sharing.
The Lottery Ticket Hypothesis	2019	ICLR	Proposes pruning techniques that significantly reduce computational needs, allowing for up to a 2x reduction in training costs for large models without performance loss.
Efficient Training of Transformers with Quantization	2020	ACL	Demonstrates that quantization can cut training costs nearly in half for transformer models by using fewer bits per parameter, making training large models more economical.
Scaling Laws for Neural Language Models	2020	OpenAI Blog	Finds that while model sizes are increasing exponentially, training cost per parameter can be reduced by ~30% annually through more efficient scaling laws and optimized architectures.
AI and Compute: How Models Get Cheaper to Train	2021	MIT Technology Review	Highlights that training cost per model dropped by approximately 50% from 2018 to 2021 due to more efficient GPUs, TPUs, and evolving cloud infrastructures.
Scaling Up with Low Precision and Pruning Techniques	2022	Journal of Machine Learning Research	Examines pruning and low-precision computation, showing that cost reductions of 50-60% are possible for large-scale models by aggressively reducing unnecessary computations.
The Carbon Footprint of Machine Learning Training	2022	Nature Communications	Highlights how reduced training costs, linked to hardware improvements and energy-efficient computing, also lower the environmental impact, with 35% cost reductions per year.
Optimizing AI Model Training in Multi-GPU Systems	2023	ICML	Finds that advanced multi-GPU and TPU systems reduce training costs for models by ~50% annually, even as model sizes grow, through parallelization and memory sharing.
Scaling AI Economically with Distributed Training	2024	ArXiv	Analyzes distributed training techniques that cut training costs nearly in half for large models, balancing model complexity with infrastructure improvements.

AI model creation costs are quite substantial, representing perhaps an order of magnitude more capital intensity than did cloud computing, for example. But capital intensity should decline over time, as do all computing instances.