Dragonfly Managing Partner Haseeb says AI extinction risk should not be voluntarily assumed before its elimination is confirmed
BIBIBI
AT A GLANCE
Dragonfly Managing Partner Haseeb believes that AI the risk may be overestimated, but one should not take risks before extinction risk is confirmed to have been eliminated.
Article
Dragonfly Managing Partner:AI Risk may be overestimated, but “risk of ruin” must never be touched
PANews September 25 reports that a Dragonfly Managing Partner wrote that if someone believes that AI the probability of causing human extinction (P(doom)) is 10%, this effectively means they believe that 90%there would be no serious consequences under those circumstances, but humanity cannot determine which circumstance it is actually in. Therefore, humanity should not voluntarily assume this risk until it can confirm that the extinction risk has been eliminated. Haseeb added that he was once a professional gambler, and that one of gambling’s core principles is never to touch “risk of ruin” (risk of ruin), meaning one must always retain chips to rebuild and deal with the worst-case scenario.AI The risk may not be as serious as imagined and may be easy to resolve, but as long as a reasonable possibility remains, one should not take risks without sufficient confirmation.
PANews reports that a Dragonfly Managing Partner published a post discussing AI the risk of causing human extinction.
02
The article uses 10%'s AI probability of extinction as an example and notes that this also means 90%there would be no serious consequences under those circumstances.
03
Haseeb stated that humanity cannot determine which circumstance it is actually in and therefore should not voluntarily assume risk before confirming that extinction risk has been eliminated.
04
Haseeb said he was once a professional gambler and explained “risk of ruin” as always retaining chips for rebuilding and dealing with the worst-case scenario.
05
Haseeb believes AI the risk may not be as serious as imagined and may be easy to resolve, but as long as a reasonable possibility remains, one should not take risks without sufficient confirmation.
AI-assisted interpretation
The following is analysis, separate from reported facts. Verify important claims independently.
This is a risk-management view: even if the probability of the worst outcome is low, if its consequences could be irreversible, one should first preserve a margin of safety rather than wager all chips on an optimistic outcome.
Why it matters to readers
This view emphasizes that, when facing AI risks that could cause extinction, avoiding irreversible consequences should take priority, and sufficient confirmation should be obtained before taking risky actions.
Beginners may understand “risk of ruin” as: do not let a single error in judgment produce an outcome from which recovery is impossible. This article expresses a personal risk assessment, not scientific proof of AI the probability of extinction.
Risks and unknowns
10%The figure in the article is a hypothetical example and cannot be used to confirm AI the actual probability of causing human extinction.
The original article provides no research evidence supporting or refuting AI extinction risk.
The original article does not explain how to determine that extinction risk has been eliminated, nor does it propose specific governance measures.
The content was relayed by PANews and does not include Haseeb’s original post.
Related Developments
Loading event timeline…
Related concepts
P(doom)
This term is not in the glossary yet. Browse related concepts in the glossary.
