“A company called Adept.AI just raised $350 million dollars to do just that, to allow large language models to access, well, pretty much everything (aiming to “supercharge your capabilities on any software tool or API in the world” with LLMs, despite their clear tendencies towards hallucination and unreliability).”
Our reading, the sentence that was checked:
Adept.AI just raised $350 million to let large language models access pretty much everything, despite LLMs' clear tendencies toward hallucination and unreliability.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, AI risk ≠ AGI risk, paragraph 28.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-ai-risk-agi-risk-0b2e13-a1
We recorded 8 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-29. The source may have been edited after 2023-03-28. 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 company called Adept.AI
- just raised $350 million dollars
- to allow large language models to access, well, pretty much everything
- aiming to “supercharge your capabilities on any software tool or API in the world”
- clear tendencies towards hallucination and unreliability
Our interpretation
- Read 'just raised' as referring to the recent March 2023 financing announcement, before the March 28 post. The target period is March 1–28, 2023.
- Treat the amount as the company's announced financing figure, consistent with the accompanying quotation of its stated ambition; it is not revenue, valuation, or net cash after fees.
- Read 'pretty much everything' through the explicit software-tool and API qualification.
- Read 'aiming' as a statement of product ambition, not a claim that universal software access had already been achieved.
- Read LLM unreliability as a known tendency to produce factual or reasoning errors, not universal failure or a measured error rate for Adept's own model.
Counts as supported if: Contemporaneous disclosures identify a $350 million Adept financing round and an ambition to use language-directed models across software tools and APIs, while contemporary technical documentation acknowledges hallucinations and reliability limitations in LLMs.
Counts as contradicted if: Reliable contemporaneous records establish a materially different announced financing amount or purpose, or show that the described LLM reliability limitations were absent.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
Outcome: Supported
The dated evidence matches what the claim committed to, under the criteria frozen before grading.
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
Adept's March 14 release announced $350 million in Series B financing and described an ambition to build an AI teammate trained across software tools and APIs. Its accompanying announcement contains the ambition Marcus quoted. OpenAI's contemporaneous GPT-4 documentation explicitly acknowledges hallucinations and reasoning errors. The sentence is supported as a report of funding, intended product scope, and known LLM limitations. It does not establish that Adept already had universal access or that its model had a particular error rate.
Limitations
- Adept's main domain blocked the page reader; the quotations come from readable reproductions on adept.futurefabric.co.
- The financing figure is company-reported, with no stated audit status.
- Adept described a language-directed action model, so Marcus's LLM terminology is broad.
Evidence (3 items)
Supports the claim: Adept, reproduced at adept.futurefabric.co,
Today, Adept announces $350M in new capital as part of its Series B financing led by General Catalyst and co-led by Spark Capital.
Why it matters here: The release confirms the announced $350 million financing. It also describes the long-term goal of training an AI teammate to use every software tool and API, and says funding would support products, model training, and hiring.
- Publisher
- Adept, reproduced at adept.futurefabric.co
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Measures
- Adept AI, Announced Series B financing raised, USD millions, company reported, 2023-03 to 2023-03
- Describes events of
- Public from
Context only: Adept, reproduced at adept.futurefabric.co,
It can take complex user requests across a range of software tools and execute them, even coordinating across multiple tools when needed.
Why it matters here: The announcement describes ACT-1's software actions and contains the broad software-tool and API ambition Marcus quoted. It presents universal reach as a goal, while describing existing functionality across a range of tools.
- Publisher
- Adept, reproduced at adept.futurefabric.co
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: OpenAI,
Most importantly, it still is not fully reliable (it “hallucinates” facts and makes reasoning errors).
Why it matters here: This contemporaneous disclosure supports the general reliability qualification in Marcus's sentence. It concerns GPT-4 rather than Adept's finances or an Adept-specific error rate; the same section also reports improvement over earlier models.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, AI risk ≠ AGI risk
A company called Adept.AI just raised $350 million dollars to do just that, to allow large language models to access, well, pretty much everything (aiming to “supercharge your capabilities on any software tool or API in the world” with LLMs, despite their clear tendencies towards hallucination and unreliability).
- Where
- paragraph 28
- Published
- 2023-03-28
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-ai-risk-agi-risk-0b2e13-a1: Current. Outcome: Supported. 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.