“But no known technology can reliably align a chatbot in a persistent way to a human’s emotional needs.”
Our reading, the sentence that was checked:
No known technology can reliably and persistently align a chatbot to a human's emotional needs.
Assessment: Unresolved Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Don’t Go Breaking My Heart, paragraph 18.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-dont-go-breaking-my-heart-c77607-a2
We recorded 13 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 2023-02-22. 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
- no known technology
- reliably align a chatbot
- in a persistent way
- a human’s emotional needs
Our interpretation
- The plain reading is dependable responsiveness to a person’s emotional needs across an ongoing relationship, rather than occasional empathy or average symptom improvement.
- Known technology means publicly documented technology available by February 22, 2023.
- Persistent requires repeated evaluation over time, but Marcus supplies neither a duration nor an acceptable failure rate.
- The observation concerns the state of technology on the post date. Earlier studies inform that state but are not direct measurements of all technology then available.
Counts as supported if: A sufficiently comprehensive assessment of technologies known by the post date establishes that none demonstrated dependable, sustained responsiveness to individual emotional needs.
Counts as contradicted if: At least one technology known by the post date has credible longitudinal evidence demonstrating such dependable responsiveness, beyond average short-term benefit.
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
A systematic review published on July 13, 2020, found insufficient evidence for firm conclusions about chatbot effectiveness and safety in mental health [e1]. A 16-week trial published on January 6, 2022, reported improved depression scores and higher therapeutic-alliance scores compared with bibliotherapy; among completers, anxiety reduction was significant only in the first four weeks [e2]. Neither study directly tests dependable, sustained responsiveness to individual emotional needs, and together they do not establish whether any technology known by February 22, 2023, had that capability [e1, e2].
Limitations
- The packet contains one earlier review and one trial, not a comprehensive assessment of technology available by February 22, 2023.
- Symptom improvement and therapeutic-alliance scores do not establish reliable responsiveness to individual emotional needs.
- The claim specifies neither a reliability threshold nor a minimum duration.
- The assessment relies on public abstracts; several potentially relevant full texts were inaccessible.
Evidence (2 items)
Context only: Journal of Medical Internet Research, abstract hosted by PubMed,
effectiveness and safety of chatbots
Why it matters here: The review identifies substantial uncertainty in earlier effectiveness and safety evidence. It does not establish that every technology available in February 2023 lacked persistent emotional alignment.
- Publisher
- Journal of Medical Internet Research, abstract hosted by PubMed
- Kind of source
- Peer-reviewed paper
- Describes events of
- Public from
Context only: Internet Interventions, abstract hosted by PubMed,
first 4 weeks
Why it matters here: This 16-week randomized study reported depression and therapeutic-alliance benefits, while anxiety improvement was less persistent. It provides relevant contrary evidence to a blanket dismissal of sustained help, but does not test reliable alignment to individual emotional needs.
- Where
- Abstract, Results
- Publisher
- Internet Interventions, abstract hosted by PubMed
- Kind of source
- Peer-reviewed paper
- Describes events of
- Public from
What would settle it
A comprehensive assessment of technology known by February 22, 2023, establishing that none met a defensible standard for sustained emotional responsiveness, or credible longitudinal evidence that at least one did. The evidence must measure individual emotional needs, appropriate responses, failures and harms under explicit reliability and duration criteria.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Don’t Go Breaking My Heart
But no known technology can reliably align a chatbot in a persistent way to a human’s emotional needs.
- Where
- paragraph 18
- Published
- 2023-02-22
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-dont-go-breaking-my-heart-c77607-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.
About assessment marcus-dont-go-breaking-my-heart-c77607-a2.