“The constant refrain I hear from VCs and AI fantasists is that "chips will bring down the cost of inference," yet I don't see any proof of that happening, nor do I think it'll happen quickly enough for these companies to turn things around.”
Our reading, the sentence that was checked:
Chips will not bring down inference costs quickly enough for AI companies to turn their finances around.
Assessment: Unresolved No deadline stated on 2026-10-29 Evidence as of
Where and when Ed Zitron wrote it
Where's Your Ed At, Godot Isn't Making it, paragraph 61.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment wyea-godot-isnt-making-it-4278f1-a2
We recorded 33 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-28. The source may have been edited after 2024-12-03. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Forecast
- Deadline, as the speaker wrote it
- None stated. This site does not add one.
- Deadline state when assessed
- No deadline stated
- Deadline state on 2026-10-29
- No deadline stated (the release’s reference date)
From the speaker
- chips will bring down the cost of inference
- nor do I think it'll happen quickly enough
- for these companies to turn things around
Our interpretation
- These companies refers principally to OpenAI and Anthropic in the surrounding passage.
- Turn things around means reaching sustainable financial viability before financial failure.
- The forecast concerns whether chip-driven savings arrive in time to change company finances, not merely whether token prices or hardware throughput improve.
- No fixed deadline or numerical profitability threshold is supplied.
Counts as supported if: Terminal financial failure of the relevant companies despite access to improved inference hardware, with evidence that chip-driven savings were insufficient or arrived too late.
Counts as contradicted if: Sustained financial recovery at the relevant companies, with evidence that chip-driven inference savings materially enabled recovery before financial failure.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Did the forecast come true?
Outcome: Unresolved
The record was checked, and the evidence available by the cutoff does not settle it.
Scored on a plain reading. The speaker named no measure for this claim, so it was judged on the reading an informed reader would take, stated below under our interpretation. Both reviews had to find that reading fair.
- Evidence as of
- Deadline state when assessed
- No deadline stated
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
NVIDIA reported on April 2, 2025 that an eight-GPU B200 system achieved 3.1 times the throughput of its eight-GPU H200 submission on the Llama 2 70B Interactive benchmark [e1]. Anthropic's filing, available October 26, 2026, reported $974 million in operating income for the second quarter and $3.929 billion in operating cash inflow for the first half of 2026 [e2]. These results show benchmark gains and financial improvement, but do not establish lasting recovery materially enabled by chip-driven inference savings or terminal failure showing those savings were insufficient [e1, e2]. The forecast remains unresolved.
Limitations
- Benchmark throughput does not measure either company's production costs or separate hardware savings from software improvements.
- Anthropic's operating income and operating cash flow do not establish net profitability or lasting financial viability.
- The packet provides no comparable OpenAI financial evidence and no causal reconciliation linking chip savings to either company's financial outcome.
Evidence (2 items)
Context only: NVIDIA,
B200 also delivered significant speedups
Why it matters here: Measured throughput improvements make inference savings plausible and weigh against any absolute assertion of no hardware progress. They do not measure OpenAI's or Anthropic's actual costs or establish financial recovery.
- Publisher
- NVIDIA
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: Anthropic, PBC / U.S. Securities and Exchange Commission,
credible trajectory
Why it matters here: The filing reports $974 million of second-quarter operating income and $3.929 billion of first-half operating cash inflow, material evidence of financial improvement. It also reports substantial net losses and explains working-capital and financing effects; these figures do not isolate chip savings or establish lasting viability.
- Publisher
- Anthropic, PBC / U.S. Securities and Exchange Commission
- Kind of source
- Regulatory filing, unaudited
- Measures
- Anthropic, Operating income and operating cash flow, USD millions, gaap, 2026-01-01 to 2026-06-30
- Describes events of
- Public from
What would settle it
Financial records for OpenAI and Anthropic establishing lasting recovery or terminal financial failure, together with comparable production-cost evidence isolating chip-driven savings and showing whether they materially enabled recovery or arrived too late or were insufficient.
Where it was said (1)
Said once in the sources this register covers.
: Where's Your Ed At, Godot Isn't Making it
The constant refrain I hear from VCs and AI fantasists is that "chips will bring down the cost of inference," yet I don't see any proof of that happening, nor do I think it'll happen quickly enough for these companies to turn things around.
- Where
- paragraph 61
- Published
- 2024-12-03
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment wyea-godot-isnt-making-it-4278f1-a2: Current. Outcome: Unresolved. Evidence as of 2026-10-28. Public date not recorded.
Recorded events
Assessment publishedBy this site
First assessment published after both reviews and the final-text check passed.