Recent Posts


Billionaires are leaving the room with excuses — Bezos says “AI will solve climate crisis”

Dystopian crash. Fantasy in ruins.

Image by Eynoxart from Pixabay

By Jo Nova

Back in 2020 nothing was more important than climate change. Swept away by the crisis, Amazon Founder Jeff Bezos set up a $10,000 million dollar fund to ‘fight climate change and protect nature’.  It was the largest piece of atmospheric philanthropy known to mankind. But it’s become a bit of a ghost account.

The money was supposed to be spent by 2030, but six years in, 72% of the funds still haven’t been allocated.

Bezos Earth Fund Is Off Pace to Meet $10 Billion Climate Giving Pledge

By Ben Elgin and Sophie Alexander, Bloomberg

Officials at the Bezos Earth Fund said they remain committed to disbursing $10 billion by 2030. With four years left on the timeline, however, only 28% of the promised funds have been allocated so far.

In interviews and statements, Earth Fund executives reaffirmed their plan to spend $10 billion by the end of the decade. “Many of our strategies remain the same,” a spokesperson said…

Now that Trump sucked the government subsidies away from pet climate projects, you’d think the Bezos funding would be more “needed” than ever.

It’s not just Bezos who is backing away slowly

Bill Gates launched “Breakthrough Energy’s Catalyst Fund” in 2021. But last  October he started talking about how climate change was not such a big deal, and they should be more focused on real issues that matter to the poor.
In February the team announced they would be sacking staff and stopping new investment:

Bill Gates’ Breakthrough Energy Catalyst Halts New Investment, Lays Off Staff

Breakthrough Energy, the Bill Gates-backed group that bankrolls climate tech, has stopped making new investments from one of its key funds to scale up nascent green technologies, just as the market for these startups is being squeezed by President Donald Trump’s assaults on climate policy.

After supporting 10 startups and spending “high hundreds of millions of dollars,” a Breakthrough Energy spokesperson said, it has now suspended new investment from the fund.

They all have their excuses for the turnaround

Jeff Bezos says AI will solve climate change. Even though AI is consuming vast quantities of gigawatts, pushing old coal plants back into action, it will undo the crisis about climate change (especially if it accidentally looks at the scientific data, eh?)

The Bezos Earth Fund believes AI will be a solution to the climate crisis.

By Elike Roohi, Alliance

According to the Fund, it is ‘harnessing [AI’s] power to transform how the world addresses climate and nature’.

Dr Stephan Singer, a senior global energy policy adviser with Climate Action Network International, said in a report by the Guardian in 2024: ‘The projects of the Bezos fund do not address the key issues of the fundamental climate crisis we are facing – they are nice but unfortunately cosmetic.’

Dr Singer was discussing the Fund’s big bet on carbon offsetting schemes, widely understood to do little to mitigate global heating.

Bill Gates says technology has already saved us

Gates was thrilled at how well the decarbonization plan was going. If we pretend that giant unaccountable organizations like the IEA make useful predictions, rather than hyperbolic spam, then things are going much better than they might have been in the IEA doomer imaginary future.

Bill Gates Highlights Massive 40% Drop in Projected CO₂ Emissions

In a LinkedIn post published yesterday, Microsoft co-founder and climate investor Bill Gates shared an optimistic update on global energy-related CO₂ emissions that challenges much of the prevailing doom-and-gloom narrative.

The key graphic Gates posted tells the story clearly:

      • In 2014, the International Energy Agency (IEA) projected that, under then-current policies, the world would be emitting roughly 50 billion tons (50 Gt) of energy-related CO₂ per year by 2040.
      • The IEA’s latest 2024 assessment, using the same “current policies” framework, now puts that 2040 figure at just 29 Gt — a stunning 42% reduction in the expected emissions trajectory in only one decade of forecasting updates.

The 2020s was the ‘decisive decade’, said Bezos. Indeed.

They’ll never say the skeptics were right.

 

10 out of 10 based on 10 ratings

11 comments to Billionaires are leaving the room with excuses — Bezos says “AI will solve climate crisis”

  • #
    Honk R Smith

    Allow me to recap.
    We’re all gonna die unless we Net Zero by next week.
    But we need massive energy for Data Centers for AI so it turns out we can’t by next week.
    Don’t worry … not a problem.
    We’ll use the AI to save us.
    Because we know the magical promise of AI will be just like the magical promise of free renewable energy from the Sun and wind.

    100

    • #
      Just Thinkin'

      Question.

      “Is AI plagiarism?”

      Mind you, I call it AB; Artificial Bull*hit.

      Their ” information” still needs to be FACT-Checked.

      10

  • #
    David Maddison

    Don’t they just have to build more windmills and solar panels to provide “free” (sic) power for all the AI?

    10

  • #
    Ronin

    It must be hard to unwind all those flaky promises to ‘save the world’, Bezos probably needs the money to keep blowing up his rockets.

    10

  • #
    David Maddison

    I asked Gulag AI:

    Knowing that enormous amounts of cheap power are needed to run AI data centres, would you as an AI recommend running exclusively on wind and solar?

    It said:

    Running AI data centers exclusively on wind and solar is technically challenging because these grids require enormous, continuous base-load power and have fluctuating outputs. While major tech companies are heavily investing in renewable energy, a completely green, 100% wind-and-solar grid often requires massive advancements in battery storage or supplemental nuclear, geothermal, or hydro power to prevent downtime.

    Leading tech and energy firms are exploring several strategies to balance clean energy with high demands:

    Next-Generation Nuclear: Companies like Microsoft and Amazon are partnering with nuclear energy providers to secure steady, carbon-free baseloads.

    Advanced Battery Storage: Mega-scale lithium-ion and alternative storage solutions (like iron-air or thermal storage) are being deployed to capture excess solar and wind for use during calm or dark periods.

    Geographic Load Shifting: AI workloads can be dynamically routed to data centers located in regions where renewable generation is peaking at that moment.

    Ultimately, while wind and solar are foundational to a zero-carbon future, an energy mix is generally the most realistic path to sustain the heavy, 24/7 demands of modern AI infrastructure today.

    So even fully woke Gulag admits the problem.

    10

    • #
      David Maddison

      I also asked Grok, which is less woke (Australia gets a special mention):

      No, I would not recommend running AI data centers (or the grid powering them) exclusively on wind and solar.

      Why reliability and cost matter for AI

      AI training and inference are extremely power-hungry and benefit from 24/7 dispatchable electricity. Data centers aim for very high uptime (often 99.99%+).

      Intermittent sources create fundamental mismatches:

      Wind and solar output varies with weather, time of day, and season. Capacity factors are typically 25-40% for solar and 30-50% for wind in good locations — far below the 80-95%+ for nuclear, hydro, or combined-cycle gas. You need massive overbuilding + storage or backup to achieve reliable supply.

      Batteries (lithium, etc.) are currently too expensive for multi-day/seasonal storage at grid scale. They help with short-term smoothing but don’t solve the “Dunkelflaute” (prolonged low-wind/low-sun periods) common in many regions.

      Backup is almost always required today: natural gas, coal, or hydro. This makes a “100% wind + solar” system far more expensive than headline LCOE (levelized cost of energy) numbers suggest. Real-world system costs (including transmission, balancing, and firming) are substantially higher.

      Empirical track record

      Places pushing hard for near-exclusive wind/solar (e.g., California, Germany, parts of Australia) show:

      Rising electricity prices and reliability issues despite heavy subsidies.

      Continued reliance on fossil backups.

      High curtailment (wasted renewable output) when generation exceeds demand/transmission.

      AI-specific demand is growing explosively (hundreds of MW to GW-scale clusters). Hyperscalers like Microsoft, Google, and Amazon have been signing deals for nuclear restarts, small modular reactors (SMRs), and long-term gas because renewables alone don’t deliver the needed firm power on timelines that match AI growth.

      Cheap, abundant energy is the constraint on AI progress. Restricting to intermittent sources would raise costs, slow deployment, and reduce global competitiveness.

      Better practical approaches

      A pragmatic mix for AI data centers prioritizes:

      Nuclear (existing plants, restarts, SMRs): High capacity factor, low emissions, dense, fuel-secure. Excellent match for always-on compute.

      Geothermal (where viable) and hydro.

      Natural gas as flexible bridge/backup — lower emissions than coal, quick ramp.

      Wind + solar as supplements where they are cheapest on the margin (good resources + transmission). They can lower average costs but need firming.

      Long-term: advanced nuclear, enhanced geothermal, and potentially fusion if it scales.

      Over-reliance on weather-dependent sources increases vulnerability to supply shocks. Historical energy transitions show abundance comes from high-energy-density sources (coal → oil/gas → nuclear). Solar/wind are useful but dilute and variable.

      Bottom line

      Exclusively wind and solar would be counterproductive for AI data centers. It would make power more expensive and less reliable precisely when we need the opposite to accelerate intelligence and scientific discovery. The physics and economics favor a diverse, high-uptime portfolio with a strong role for firm, scalable sources like nuclear. Policy that forces exclusivity (mandates without storage reality) raises costs for everyone. Maximize total clean, cheap, reliable energy production instead.

      10

      • #
        Greg in NZ

        Or, as our parents reiterated their parents’ advice:

        Do NOT put all your eggs in one basket.

        Mind you, who can afford eggs these days?!?

        10

    • #
      Dave in the States

      Well the AI dept. at gulag can say it. Regular employees better not.

      these grids require enormous, continuous base-load power

      Doesn’t traditional industry, hospitals, schools, rock-n-roll bands, laundry mats, resturants, and so forth, also need cheap continuous base load power?

      20

      • #
        Greg in NZ

        Sir, that type of thinking is known as COMMON SENSE which is not only not allowed, it’s as rare as hen’s teeth. Now where did I put those eggs…

        10

  • #

    A good friend of mine (we NEVER talk politics) is terrified of AI because of it’s power munching and thirst for water. He also said “well done” to me when I changed my diet away from 500 gm steaks to 200 gm steaks because it’s good for the planet. Please.
    Read a History book lads ‘n’ ladies. Every Malthusian prediction so far’s had a use by date. The 20th (and early 21st) Century has seen a spike in Malthusian BS thanks to mass media. I told some idiot on FB the other day that Ehrlich was wrong and his retort was “No, you’re wrong.” I mean how (I mean why) do you argue with that? People are so stupid!
    Some other idiot I told off was whining that we’re building a new stadium in Brisbane. She said “we should be building a new hospital!” I asked her how a hospital would grow the pie? No answer.
    OK, so I’ve just introduced a three pronged argument that I’d have thought any sane reader could understand: shut up, think critically, and get to work. But no, I have been accused of having a low IQ or something. Mmmm, I wish I was stupid. Stupid people are happier I tell myself, but then the three morons above are stupid and deeply unhappy. What a race we are.

    30

  • #
    Forrest Gardener

    And once more grok produces evidence that it will say anything you want it to say. But only after being misleading and deceptive.

    After a brief discussion about whether the $10B by Bezos was philanthropic or commercial…

    Q: so bezos would not personally or through one of the entities he controls gain any intellectual property or profit at all

    A: No, it’s not accurate to say there is zero possibility of any benefit or IP access to Bezos or entities he controls.

    Bottom line: This is structured as philanthropy, not a for-profit investment vehicle. Bezos does not get direct financial returns or exclusive IP ownership. However, certain grant programs include licenses that provide affiliates (potentially including business interests connected to Bezos) with broad usage rights — which some recipients and universities see as a non-trivial advantage. It’s not pure “no benefit whatsoever,” but far from a disguised profit scheme.

    00

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>