Pundit ClaimsAI claims by Gary Marcus and Ed Zitron, with the evidence

“Sure you can say “it’s projections,” but there is a clear attempt to use the media to misinform investors and the general public. For example, OpenAI claimed SoftBank would spend $3bn a year on agents in 2025. That never happened!”

The AI Bubble Is An Information War, paragraph 68. The full statement is at the source.

Our reading, the sentence that was checked:

There is a clear attempt by OpenAI to use the media to misinform investors and the public; for example, OpenAI claimed SoftBank would spend $3bn a year on agents in 2025, which never happened.

Assessment: Unresolved Evidence as of

Where and when Ed Zitron wrote it

Where's Your Ed At, The AI Bubble Is An Information War, paragraph 68.

Said
Source observed
live page, full text, public
Evidence as of
the cutoff for this assessment
Assessment recorded
assessment wyea-the-ai-bubble-is-an-information-war-068ddf-a1

We recorded 38 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 2026-03-03. This page does not claim the wording was captured then.

What the claim commits to

Kind
Observation
Deadline, as the speaker wrote it
None stated. This site does not add one.
Deadline state when assessed
No horizon
Deadline state on 2026-10-29
No horizon (the release’s reference date)

From the speaker

  • a clear attempt to use the media to misinform investors and the general public
  • OpenAI claimed SoftBank would spend $3bn a year on agents in 2025
  • That never happened!

Our interpretation

  • Assess both the cited spending example and the allegation that it demonstrates intentional misinformation.
  • Read 'in 2025' as calendar 2025, rather than SoftBank's fiscal year or a future annual spending rate.
  • Read 'never happened' as failure to spend approximately $3 billion during that year, not proof that no OpenAI products were used.
  • A missed projection alone does not establish that its maker intended to deceive.

Counts as supported if: A documented commitment to approximately $3 billion of agent spending during calendar 2025, evidence that the spending did not occur, and evidence connecting the representation to an attempt to mislead.

Counts as contradicted if: Comparable records establish that the promised 2025 spending occurred, or establish that the attributed calendar-2025 promise was materially misstated. Lack of evidence of intent alone is not contradiction.

Retrospective: these criteria were written after the outcome was public.

Current assessment

Was it true for the period described?

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 horizon
Assessment recorded
Policy version
2026-10-28.7

Rationale

The annual $3 billion plan was real, but the original announcement covered broader OpenAI solutions and did not clearly require $3 billion of spending in calendar 2025. SoftBank's CFO said payment timing remained undetermined. The November announcement planned customer availability for 2026, but that does not establish actual 2025 spending. No inspected record proves the claimed spending failure or intentional misinformation. A missed projection would not, by itself, establish deception.

Limitations

  • Rollout timing cannot substitute for payment records.
  • SoftBank Corp.'s fiscal-year disclosures do not directly answer a calendar-year claim about the entire group.

Evidence (3 items)

  • Context only: OpenAI and SoftBank Group,

    a partnership

    Why it matters here: The joint announcement includes a $3 billion annual plan covering OpenAI solutions, including ChatGPT Enterprise. Its spending provision does not explicitly promise a full $3 billion of realized spending during calendar 2025.

    Publisher
    OpenAI and SoftBank Group
    Kind of source
    Company disclosure: the company describing itself, not an independent finding
    Measures
    SoftBank Group and its group companies, Announced annual spending on OpenAI solutions, USD billions per year, company reported, ? to ?
    Describes events of
    Public from
  • Context only: SoftBank Group,

    has not yet been determined.

    Why it matters here: The CFO expressly left payment timing undetermined and linked payments to completion and launch of the service. This materially qualifies the attribution of a firm calendar-2025 promise.

    Publisher
    SoftBank Group
    Kind of source
    Company disclosure: the company describing itself, not an independent finding
    Measures
    SoftBank Group, Timing of the announced OpenAI service payment, USD billions, company reported, ? to ?
    Describes events of
    Public from
  • Context only: SoftBank Group, SoftBank Corp. and OpenAI,

    availability planned for 2026.

    Why it matters here: Customer availability was planned for 2026, although internal adoption was to precede that rollout. Product timing does not establish the amount paid during 2025.

    Publisher
    SoftBank Group, SoftBank Corp. and OpenAI
    Kind of source
    Company disclosure: the company describing itself, not an independent finding
    Describes events of
    Public from

What would settle it

The agreement's effective spending obligations and complete calendar-2025 service payments or expenses, together with contemporaneous evidence bearing on the alleged intent to mislead.

Where it was said (1)

Said once in the sources this register covers.

  • : Where's Your Ed At, The AI Bubble Is An Information War

    Sure you can say “it’s projections,” but there is a clear attempt to use the media to misinform investors and the general public. For example, OpenAI claimed SoftBank would spend $3bn a year on agents in 2025. That never happened!

    Published
    2026-03-03
    Source observed
    , live page
    Access
    full text, public

History

Assessment versions

Recorded events

  1. Assessment publishedBy this site

    First assessment published after both reviews and the final-text check passed.

    About assessment wyea-the-ai-bubble-is-an-information-war-068ddf-a1.