Thesis · Foundation Models & AI Intelligence

Will US data center electricity consumption, as reported by EIA, grow at least 15% year-over-year in the most recent full-year or twelve-month figure published before 2027-12-31?

Elastic inference demand translates into physical compute expansion, so data center electricity consumption keeps growing double-digit percentages despite per-token cost declines.

6 dated tests · 0 resolved · next on

The tests More theses in Foundation Models & AI Intelligence

Who is on each side

The two sides.

Each side in DeepStack’s words, then the voices who argue it, attributed and paraphrased, dated where the date is known. A voice links to its page on DeepStack where one exists.

The case for yes

Inference demand is price-elastic

Cheaper inference unlocks agentic workloads consuming 15-100x more compute per task, driving utilization of deployed accelerators toward saturation. Total electricity drawn by US data centers rises faster than efficiency gains reduce energy per token.

The case for no

The case against it

Market preference shifting from high-cost to lower-cost token models, so silicon Data Token Cost Index declines while overall AI spending remains stable or increases. It would show first if EIA-reported data center load growth falls below 15% YoY while token prices continue to fall.

6 voices

Inference demand is price-elastic

  1. Invest in Assets

    Cheaper inference unlocks use cases that were previously uneconomical, and the computational intensity of agent workloads multiplies demand even as the price per token falls, expanding the total addressable market and maintaining or growing revenue…

    Source

  2. Arctan Research

    When an input gets cheaper, demand can expand enough that total outlay on the input rises; applied to AI, cheaper models would speed adoption and lift chip demand, provided the leading chip designer keeps its relative advantage. Because the effect…

    Source

  3. Kevin Gargate Osorio

    When AI becomes cheaper to develop and deploy, more users and companies adopt it, increasing overall infrastructure demand rather than reducing it, following Jevons' paradox.

    Source

  4. Alex Dc

    As AI technology becomes more accessible and cheaper to run, adoption accelerates across more users and applications, multiplying the total resource consumption even as per-unit costs fall, preserving and expanding the market for chip suppliers…

    Source

  5. Timothy B Lee

    When the cost of individual computation falls while quality improves with more computation, users increase token generation to solve problems better. This happened historically with computing power, demand rose as costs fell, and should repeat with…

    Source

  6. Kris

    When technology makes GPU-based AI more efficient and cheaper, adoption expands to new applications and users previously unable to afford it, partially offsetting the per-unit efficiency gain and driving higher absolute demand for chips despite…

    Source

1 voice

The case against it

  1. Moses Sternstein

    If demand for lower-cost tokens increases more rapidly than demand for high-cost tokens, the token-spend intensity index falls as a mathematical consequence, even though total token consumption and spending continue growing. This would occur as…

    Source

What would settle it

The dated tests.

Each test names the public source that settles it, the rule and the day. The Call’s number, the readers’ median and the machines’ numbers stand beside it, as on the Docket.

  1. Alphabet, revenue growth from a year earlier, quarter ended September 30, 2026: whether it comes in at or above 22%.

    On the Docket Pendingth092-t04Pre-specified

    Question
    Did the revenue of Alphabet for the quarter ended September 30, 2026 grow at or above 22.0% from the quarter ended September 30, 2025?
    Source
    SEC EDGAR, XBRL company facts of Alphabet (Forms 10-Q and 10-K)
    Rule
    Compare the revenue that Alphabet reports for the quarter ended September 30, 2026 with the quarter ended September 30, 2025, in its 10-Q or 10-K filing on SEC EDGAR. The answer is yes if growth is at or above 22.0%; no if it is below 22.0%.
    If the source is late, revised or silent
    If Alphabet has not filed by November 16, 2026, or a filing restates a period the rule reads, the test is inconclusive, and the Result says what was read and what was missing.
    Base rate
    Alphabet's revenue growth from a year earlier was 24.2% for the quarter ended June 30, 2026, by its filing on SEC EDGAR.
    Outside view
    The year-earlier quarter is not on the same concept in the filings, so the rungs sit on the last print alone.
    What would change the number
    The results-calendar consensus for the quarter ended September 30, 2026 implies 25.0%, and the middle rung is centred there.

    The Call, before the dateExpects yes. Gives the yes answer 72%.

    72%

    0 · no100 · yes

    The machinesClaude 22% · DeepSeek 12%

    Claude: 22%DeepSeek: 12%
    The ladder 3 rungs on one reading, Oct 28, 2026: each asks whether it lands above its threshold, so the Call’s numbers down the rungs sketch where it expects the reading to land.

    1. Above 22%Call 72% · Claude 22% · DeepSeek 12%

      72%

    2. Above 24.5%Call 50% · Claude 12% · DeepSeek 5%

      50%

    3. Above 27%Call 27% · Claude 4% · DeepSeek 4%

      27%

    0 · no50100 · yes

    Your number

  2. Alphabet, revenue growth from a year earlier, quarter ended September 30, 2026: whether it comes in at or above 24.5%.

    On the Docket Pendingth092-t05Pre-specified

    Question
    Did the revenue of Alphabet for the quarter ended September 30, 2026 grow at or above 24.5% from the quarter ended September 30, 2025?
    Source
    SEC EDGAR, XBRL company facts of Alphabet (Forms 10-Q and 10-K)
    Rule
    Compare the revenue that Alphabet reports for the quarter ended September 30, 2026 with the quarter ended September 30, 2025, in its 10-Q or 10-K filing on SEC EDGAR. The answer is yes if growth is at or above 24.5%; no if it is below 24.5%.
    If the source is late, revised or silent
    If Alphabet has not filed by November 16, 2026, or a filing restates a period the rule reads, the test is inconclusive, and the Result says what was read and what was missing.
    Base rate
    Alphabet's revenue growth from a year earlier was 24.2% for the quarter ended June 30, 2026, by its filing on SEC EDGAR.
    Outside view
    The year-earlier quarter is not on the same concept in the filings, so the rungs sit on the last print alone.
    What would change the number
    The results-calendar consensus for the quarter ended September 30, 2026 implies 25.0%, and the middle rung is centred there.

    The Call, before the dateEither answer fits the Call. Gives the yes answer 50%.

    50%

    0 · no100 · yes

    The machinesClaude 12% · DeepSeek 5%

    Claude: 12%DeepSeek: 5%

    Your number

  3. Alphabet, revenue growth from a year earlier, quarter ended September 30, 2026: whether it comes in at or above 27%.

    On the Docket Pendingth092-t06Pre-specified

    Question
    Did the revenue of Alphabet for the quarter ended September 30, 2026 grow at or above 27.0% from the quarter ended September 30, 2025?
    Source
    SEC EDGAR, XBRL company facts of Alphabet (Forms 10-Q and 10-K)
    Rule
    Compare the revenue that Alphabet reports for the quarter ended September 30, 2026 with the quarter ended September 30, 2025, in its 10-Q or 10-K filing on SEC EDGAR. The answer is yes if growth is at or above 27.0%; no if it is below 27.0%.
    If the source is late, revised or silent
    If Alphabet has not filed by November 16, 2026, or a filing restates a period the rule reads, the test is inconclusive, and the Result says what was read and what was missing.
    Base rate
    Alphabet's revenue growth from a year earlier was 24.2% for the quarter ended June 30, 2026, by its filing on SEC EDGAR.
    Outside view
    The year-earlier quarter is not on the same concept in the filings, so the rungs sit on the last print alone.
    What would change the number
    The results-calendar consensus for the quarter ended September 30, 2026 implies 25.0%, and the middle rung is centred there.

    The Call, before the dateExpects no. Gives the yes answer 27%.

    27%

    0 · no100 · yes

    The machinesClaude 4% · DeepSeek 4%

    Claude: 4%DeepSeek: 4%

    Your number

  4. Amazon, revenue growth from a year earlier, quarter ended September 30, 2026: whether it comes in at or above 12%.

    On the Docket Pendingth092-t01Pre-specified

    Question
    Did the revenue of Amazon for the quarter ended September 30, 2026 grow at or above 12.0% from the quarter ended September 30, 2025?
    Source
    SEC EDGAR, XBRL company facts of Amazon (Forms 10-Q and 10-K)
    Rule
    Compare the revenue that Amazon reports for the quarter ended September 30, 2026 with the quarter ended September 30, 2025, in its 10-Q or 10-K filing on SEC EDGAR. The answer is yes if growth is at or above 12.0%; no if it is below 12.0%.
    If the source is late, revised or silent
    If Amazon has not filed by November 16, 2026, or a filing restates a period the rule reads, the test is inconclusive, and the Result says what was read and what was missing.
    Base rate
    Amazon's revenue growth from a year earlier was 19.6% for the quarter ended June 30, 2026, by its filing on SEC EDGAR.
    Outside view
    The year-earlier quarter is not on the same concept in the filings, so the rungs sit on the last print alone.
    What would change the number
    The results-calendar consensus for the quarter ended September 30, 2026 implies 12.2%, and the middle rung is centred there.

    The Call, before the dateExpects yes. Gives the yes answer 73%.

    73%

    0 · no100 · yes

    The machinesClaude 54% · DeepSeek 63%

    Claude: 54%DeepSeek: 63%
    The ladder 3 rungs on one reading, Oct 29, 2026: each asks whether it lands above its threshold, so the Call’s numbers down the rungs sketch where it expects the reading to land.

    1. Above 12%Call 73% · Claude 54% · DeepSeek 63%

      73%

    2. Above 15%Call 46% · Claude 16% · DeepSeek 13%

      46%

    3. Above 18%Call 25% · Claude 8% · DeepSeek 3%

      25%

    0 · no50100 · yes

    Your number

  5. Amazon, revenue growth from a year earlier, quarter ended September 30, 2026: whether it comes in at or above 15%.

    On the Docket Pendingth092-t02Pre-specified

    Question
    Did the revenue of Amazon for the quarter ended September 30, 2026 grow at or above 15.0% from the quarter ended September 30, 2025?
    Source
    SEC EDGAR, XBRL company facts of Amazon (Forms 10-Q and 10-K)
    Rule
    Compare the revenue that Amazon reports for the quarter ended September 30, 2026 with the quarter ended September 30, 2025, in its 10-Q or 10-K filing on SEC EDGAR. The answer is yes if growth is at or above 15.0%; no if it is below 15.0%.
    If the source is late, revised or silent
    If Amazon has not filed by November 16, 2026, or a filing restates a period the rule reads, the test is inconclusive, and the Result says what was read and what was missing.
    Base rate
    Amazon's revenue growth from a year earlier was 19.6% for the quarter ended June 30, 2026, by its filing on SEC EDGAR.
    Outside view
    The year-earlier quarter is not on the same concept in the filings, so the rungs sit on the last print alone.
    What would change the number
    The results-calendar consensus for the quarter ended September 30, 2026 implies 12.2%, and the middle rung is centred there.

    The Call, before the dateEither answer fits the Call. Gives the yes answer 46%.

    46%

    0 · no100 · yes

    The machinesClaude 16% · DeepSeek 13%

    Claude: 16%DeepSeek: 13%

    Your number

  6. Amazon, revenue growth from a year earlier, quarter ended September 30, 2026: whether it comes in at or above 18%.

    On the Docket Pendingth092-t03Pre-specified

    Question
    Did the revenue of Amazon for the quarter ended September 30, 2026 grow at or above 18.0% from the quarter ended September 30, 2025?
    Source
    SEC EDGAR, XBRL company facts of Amazon (Forms 10-Q and 10-K)
    Rule
    Compare the revenue that Amazon reports for the quarter ended September 30, 2026 with the quarter ended September 30, 2025, in its 10-Q or 10-K filing on SEC EDGAR. The answer is yes if growth is at or above 18.0%; no if it is below 18.0%.
    If the source is late, revised or silent
    If Amazon has not filed by November 16, 2026, or a filing restates a period the rule reads, the test is inconclusive, and the Result says what was read and what was missing.
    Base rate
    Amazon's revenue growth from a year earlier was 19.6% for the quarter ended June 30, 2026, by its filing on SEC EDGAR.
    Outside view
    The year-earlier quarter is not on the same concept in the filings, so the rungs sit on the last print alone.
    What would change the number
    The results-calendar consensus for the quarter ended September 30, 2026 implies 12.2%, and the middle rung is centred there.

    The Call, before the dateExpects no. Gives the yes answer 25%.

    25%

    0 · no100 · yes

    The machinesClaude 8% · DeepSeek 3%

    Claude: 8%DeepSeek: 3%

    Your number

In The Buildout

The stories that take up this thesis, with their own Calls and tests.

A thesis page states one question and both sides of it; DeepStack takes no side in words, and its view is the number it gives each test before the date. Analysis and opinion, not investment advice. Voices are paraphrased from public writing and attributed by name; DeepStack shows no score, ranking or accuracy for any person. Pre-specified means the words were approved on or before the test’s date. Specified after the date means they were written afterwards: such a test is counted and shown, and never rated. An open question has no words yet. The Call’s number is the chance, from 0 to 100, that the Call gives the yes answer to a test’s question, recorded before the test’s date; it is DeepStack’s own reading, not a market price and not advice. The readers’ mark on the same line is the readers’ median, the middle of the numbers signed-in readers gave the test, shown from 25 readers. The machines: Each model read this page and nothing else. The number that counts is the latest one each model gave before the test closed.