Brad West🔸
Bio
Participation2
Looking to advance businesses with charities in the vast majority shareholder position. Check out my TEDx talk for why I believe Profit for Good businesses could be a profound force for good in the world.
Posts 23
Comments321
I resonate deeply with your sadness. What helps me stay anchored is identifying EA primarily as a personal commitment and life philosophy rather than merely as a movement. This perspective keeps my dedication resilient, rooted in the core EA value of boundless determination to better the world, regardless of external disruptions or individual mistakes.
Movements inevitably face setbacks and crises, but the philosophical essence of EA—its unwavering commitment to improving the world—remains solid. The movement serves as a practical tool for amplifying these core values, even if it occasionally falters.
Controversies offer opportunities to recommit individually and collectively to fundamental EA principles such as transparency, humility, and rigorous inquiry. Rather than depending solely on central figures, these moments encourage broader ownership and individual agency.
Ultimately, the enduring strength of EA lies not in flawless execution but in the earnest pursuit of doing the most good we can with the resources available. This foundational ideal, characterized by thoughtful compassion and pragmatic action, is deeply worth preserving.
Would a potential cure to the sycophancy be to reverse the framing to Claude, so that it perceives that you are your opponent and you are looking for flaws with the comment? I realize that this would not get quite what you are looking for, but getting strong arguments for the other side could be helpful.
Because we face substantial uncertainty around the eventual moral value of AIs, any small reduction in p(doom) or catastrophic outcomes—including S-risks—carries enormous expected utility. Even if delaying AI costs us a few extra years before reaping its benefits (whether enjoyed by humans, other organic species, or digital minds), that near-term loss pales in comparison to the potentially astronomical impact of preventing (or mitigating) disastrous futures or enabling far higher-value ones.
From a purely utilitarian viewpoint, the harm of a short delay is utterly dominated by the scale of possible misalignment risks and missed opportunities for ensuring the best long-term trajectory—whether for humans, other organic species, or digital minds. Consequently, it’s prudent to err on the side of delay if doing so meaningfully improves our chance of securing a safe and maximally valuable future. This would be true regardless of the substrate of consciousness.
I would like to see a strong argument for the risk of "replaceability" as a significant factor in potentially curtailing someone's counterfactual impact in what might otherwise be a high-impact job. This central idea is that the "second choice" applicant, after the person who was chosen, might have done just as well, or near just as well as the "first choice" applicant, making the counterfactual impact of the first small. I would want an analysis of the cascading impact argument: that you "free up" the second choice applicant to do other impactful work, who then "frees up" someone else, etc., and this stream of "freeing up value" mostly addresses the "replaceability concern.
Yeah I would think that we would want ASI-entities to (a) have positively valenced experienced as well as the goal of advancing their positively valenced experience (and minimizing their own negatively valenced experience) and/or (b) have the goal of advancing positive valenced experiences of other beings and minimizing negatively valenced experiences.
A lot of the discussion I hear around the importance of "getting alignment right" pertains to lock-in effects regarding suboptimal futures.
Given the probable irreversibility of the fate accompanying ASI and the potential magnitude of good and bad consequences across space and time, trying to maximize the chances of positive outcomes seems simply prudent. Perhaps some of the "messaging" of AI safety seems to be a bit human-centered, because this might be more accessible to more people. But most who have seriously considered a post-ASI world have considered the possibility of digital minds both as moral patients (capable of valenced experience) and as stewards of value and disvalue in the universe.
The preference for humans remaining alive/in control isn't necessarily speciesist because it's the qualities of having valuable conscious experience and concern for the promotion of valuable as well as avoidance of disvaluable conscious experience that might make one prefer this outcome.
We do not know whether ASI would have these qualities or preferences, but if we could know that it did, you would have a much stronger case for your argument.
I would write how there's a collective action problem regarding reading EA forum posts. People want to read interesting, informative, and impactful posts and karma is a signifier of this. So often people will not read posts, especially on topics they are not familiar, unless it has already achieved some karma threshold. Given how time-sensitive front page availability is without karma accumulation and unlikely relatively low karma posts are too be read once off the front page, it is likely that good posts could be entirely ignored. On the other hand, some early traction could result in OK posts getting very high karma because a higher volume of people have been motivated to check the post out.
I think this could be partially addressed by having volunteers, or even paying people, to commit to read posts within a certain time frame and upvote (or not, or downvote) if appropriate. It might be a better use of funds than myriad cosmetic changes.
Below is a post I wrote that I think might be such a post that was good (or at least worthy of discussion) but people probably wanted to freeride on others' early evaluation. It discusses how jobs in which the performance metrics actually used are orthogonal to many ways in which good can be done may be opportunities for significant impact.
https://forum.effectivealtruism.org/posts/78pevHteaRxekaRGk/orthogonal-impact-finding-high-leverage-good-in-unlikely
Another set of actors that would be incentivized in this would be the survey respondents, to say higher counterfactual values of first vs second choices. Saying otherwise could go against their goals of attracting more of the EA talent pool to their positions. The framing of irreplaceability for their staff also tends to lend to the prestige of their organizations and staff.
With limited applicants, especially in very specialized areas, I think there is definitely a case for a high value of first vs. second choice applicant. But I suspect that this set of survey respondents would be biased in the direction of overestimating the counterfactual impact.
I appreciate your exploration of the strategic complexity inherent in prioritizing effectiveness. A crucial aspect involves recognizing that impact often occurs in significant "chunks." Identifying key thresholds and accurately assessing their likelihood of being pivotal is essential for effective resource allocation. For instance, in farmed animal advocacy, securing cage-free commitments from major corporations can lead to disproportionate industry-wide improvements, making precise strategic targeting crucial. In these contexts, there might appear to be little impact until the critical moment. However, openly communicating these threshold calculations might inadvertently strengthen adversaries' resistance. Drawing from game theory's "madman" approach, an actor sometimes gains strategic advantage if adversaries believe it may irrationally commit excessive resources or accept high risks to achieve its goals, thus deterring aggressive opposition.
On a related semantic note, describing strategic resilience or integrating adversarial responses as "less effective" could oversimplify this nuanced issue. I would think when people say “effective” that they are talking about what best achieves one’s goals, and integrating adversarial responses would help in doing so.