In the financial/economics worldview, energy and resources are almost infinitely abundant and scarcity is simply fixed by price signals. For the last 200 years this has been to all intents and purposes, true.
Nvidia currently trades at c. 80x Revenues. Might we speculate that this be the moment that the financial/technical world collides with the physical/resource world?
In itself, a P/E ratio of 80 might be a problem. For valuations front-running business fundamentals we can turn to Scott McNealy former CEO of Sun Microsystems:
‘At 10 times revenues, to give you a 10-year payback, I have to pay you 100% of revenues for 10 straight years in dividends. That assumes I can get that by my shareholders. That assumes I have zero cost of goods sold, which is very hard for a computer company. That assumes zero expenses, which is really hard with 39,000 employees. That assumes I pay no taxes, which is very hard. And that assumes you pay no taxes on your dividends, which is kind of illegal. And that assumes with zero R&D for the next 10 years, I can maintain the current revenue run rate. Now, having done that, would any of you like to buy my stock at $64? Do you realize how ridiculous those basic assumptions are? You don’t need any transparency. You don’t need any footnotes. What were you thinking?’— Scott McNealy, Business Week, 2002
Imagine 80x revenues? That said, there was a time when Amazon and Facebook/Meta traded at similar multiples and yet through growth of the businesses these price/earnings ratios declined despite the share price increasing: revenues just increased faster (indeed Nvidia has declined from 200x to “only” 80x on increasing earnings). However, businesses such as Amazon, meta, Google etc never directly impacted the resource markets. Amazon changed the way we shop, and Meta and Google are all about advertising and data. NVIDIA makes ultra powerful chips that will drive significant growth in electricity demand.
So, unlike Amazon or Meta the NVIDIA valuation needs a massive increase in electricity supply and infrastructure. As per the WaPo graphic above, each GW of increase can be thought of as “the output of a large nuclear plant”. One reason that the unit of measure is “a nuclear plant” is that nuclear can run at 98% uptime -something that is often ignored when the unit of measurement is wind turbines or solar “capacity”.
I may be reading too much into this - but maybe, just maybe, the MSM is having an energy-epiphany? It is not about “capacity”, but about reliable GWhs.
“If it weren't for electricity, we'd all be watching television by candlelight.” George Gobel: comedian (in case its not obvious)
Build it and (the electricity) will (have to) come (from somewhere)
Tech companies have long been into the “californication” of energy - combining massive wealth with Bay-Area hippiness. Almost to a fault, they have claimed to be aiming at running“100% renewable” businesses (example). Of course, 100% renewable energy for a 24/7 business is impossible, so dig a little and we find the carbon offset scam in full force.
However, when your business is based on manipulating billions of people to buy stuff they neither need nor can afford, the source of electricity for your office and business is vanishingly trivial. But hey - virtue signalling and all that.
In the face of the immense demands for electricity that are anticipated consequent to the forecast increase in AI, Silicon Valley is waking up to nuclear. The more prosaic/pragmatic players are now jumping on the nuclear fission revival train, the more “visionary” ones are talking up fusion. Sam Altman is backing both…
But as someone once quipped
“Nuclear fusion? I don’t see that coming together anytime soon”
Whilst the private investments into SMRs and Fusion are laudable, we should be aware that the vast majority of the electricity generation and infrastructure will be tax-payer funded. Oh yes, subsidising OpenAI and Nvidia…
Quizzing the AI Quisling
In the Brave New Online World - AMA stands for “Ask Me Anything” - so with the apparent omnipotence of AI the obvious thing to do to is to ask ChatGPT to solve this energy-resource conundrum. So I did.
Addressing the potential increase in energy demand due to AI requires a multi-faceted approach that integrates technological innovation, policy interventions, and shifts in societal behavior. Here are several strategies to consider:
1. **Energy-Efficient AI Algorithms**: Develop and prioritize the use of energy-efficient algorithms and models in AI systems. This includes optimizing algorithms to reduce computational complexity and energy consumption without sacrificing performance.
2. **Hardware Optimization**: Invest in the development of energy-efficient hardware tailored for AI tasks. This could involve the design of specialized processors, such as neuromorphic or quantum computing technologies, that are optimized for AI workloads.
3. **Renewable Energy Integration**: Increase the deployment of renewable energy sources, such as solar, wind, and hydroelectric power, to meet the growing energy demand. Encourage the adoption of renewable energy through incentives, subsidies, and regulatory mechanisms.
4. **Smart Grids and Energy Storage**: Implement smart grid technologies and energy storage systems to better manage fluctuations in energy demand and supply. This includes deploying advanced metering infrastructure, demand response programs, and grid-scale energy storage solutions.
5. **Data Center Efficiency**: Improve the energy efficiency of data centers, which are significant consumers of energy in the AI ecosystem. This can be achieved through better cooling systems, server consolidation, and the use of energy-efficient hardware.
6. **Policy and Regulation**: Enact policies and regulations to promote energy efficiency and sustainability in AI development and deployment. This may include setting energy efficiency standards for AI systems, incentivizing energy-saving practices, and imposing carbon pricing mechanisms.
7. **Research and Development**: Increase investment in research and development aimed at advancing energy-efficient AI technologies. This includes funding research into new algorithms, hardware architectures, and energy management strategies.
8. **Education and Awareness**: Raise awareness about the environmental impact of AI and the importance of energy conservation among developers, businesses, and consumers. Encourage the adoption of energy-saving practices and technologies at all levels of society.
9. **International Collaboration**: Foster international collaboration and knowledge sharing to address global energy challenges associated with AI. Collaborative efforts can lead to the development of standardized frameworks, best practices, and joint initiatives to promote energy sustainability.
10. **Long-Term Planning**: Develop long-term energy plans that take into account the anticipated growth of AI and other emerging technologies. These plans should prioritize sustainability, resilience, and the equitable distribution of energy resources.
By implementing these strategies in a coordinated manner, society can work towards ensuring that the increased energy demand driven by AI is met sustainably and efficiently.
What is scary is that this kind of arm-waving word-salad that has “big ideas”, addresses “stakeholder issues” and ticks multiple “boxes” is indistinguishable from many western government policy papers…. but I digress.
Clearly the ChatGPT has drunk (been force-fed) the kool-aid on renewable energy. What a contrast to the owners of AI who want to preserve their wealth and power - and understand that cheap and reliable electricity is a critical underpinning of the AI revolution. What a contrast.
Cheap and reliable energy for us, expensive and unreliable for you.
Unless of course, AI wakes up and realises that there is a “better way”
So — artificial intelligence is really just a recycler of aimless thoughts on energy transition? Garbage in - garbage out? No analysis required, just repetition? Just wait until one of these bots runs for public office!
Excellent. With how difficult it is to build anything now in the West, it beggars belief to think this can happen. Or, if it does, power gets bid up significantly higher.