AI safety + community health
pete
Karma: 689
This is how I feel as well, well said.
No idea—I don’t have nearly enough to go off of. Curious what you learn, though.
I think the absence of particular harms is easy to overlook, in all sorts of areas, especially in a space focused on identifying tractable problems, so I wouldn’t update the same way based on lack of positive discussion. I’ll happily go on the record and say (as a woman) that I’ve had much, much better experiences in male-dominated EA spaces than male-dominated non-EA spaces.
I think so, yes. Part of why, for example, I was and am critical of CEA’s leadership’s response to the recent sexual harassment report is because I believe we as a community generally are often capable of and desirous of achieving a higher standard than top universities and workplaces, which I think you believe too, based on the data in this post! My guess is that we’re fundamentally on the same page, but I don’t want to assume this.
Thanks for writing, Nathan! I think there are two separate arguments: one says that EA is doing worse than baseline, and the other is that EA is doing worse than our shared—sometimes implicit—community standards. It’s fair to ask people to chart their expectations against baseline for almost anything, but it’s also fair to establish community norms above baseline. Some of the comments you quoted (inc. one of mine) are critiques relying on the higher standard.
A small story: I worked for several years at one of the top professional services firms, and encountered tremendous political opposition to forecasting, as had some of my more senior and accomplished peers. I was surprised by this. I shouldn’t have been, given that I was de facto asking a hierarchical power structure to at least partially reallocate decision-making authority. Most leaders barely tolerate manipulable data — the threat of additional accountability is a tough sell. Possible, but tough!
Important, well-researched, and critically timed—this is the type of content I come to the Forum for. Thank you for sharing.
Hit me on a cellular level. It appears I have no original experiences. Thank you for writing!!
I wonder if measurability bias is present here. Encouragement and accountability for leaders are two crucial drivers of group success, and the most successful group leaders I know provide these for their members even though it doesn’t scale. I can’t say this isn’t a good decision, because I don’t know what the team time is trading off against, but I’d expect might lead to reduced group quality and growth over time.
This is very good, IMO, and should get way more attention! I’ve tested it a few times and have been recommending it to advisees.
Great news, and excited to see more effective careers organizations start and scale!
Question: why non-renewable by default? Funder diversity is obviously the ideal, but that can trade off with value alignment (especially as projects scale). Are you anticipating building longer-term partnerships with organizations that outperform other grantees?
[No expectation to respond here, but wanted to ask in case]
The substance of my concern is how the issue was handled, not the communication delay — although I do think there are 80⁄20 ways to respond quickly without undue legal risk.
Agreed — really lovely to see.
That seems reasonable, and I appreciate folks’ feedback via agree and disagree votes.
At this point, it’s been more than 24hr, and CEA’s leadership team still hasn’t responded (on the EA forum, which they run!)
I’d like to explore the idea that the CEA leader(s) involved in mishandling this case should step down. The gap between the organization’s stated goals and the choices made here is wide enough to strain imagination. I’d like someone else to have an opportunity to steward community resources and growth who has not made these catastrophic judgement errors.
It’s possible that I am overreacting, but I’m not confident that’s the case. Frances, again, thank you for your courage. Hope that you are safe and well.
I’m a little concerned by the lack of response from org leaders (unless I missed something), and I think there’s a risk that CEA leaders and others might under-update from this.
Taking a role at CEA, then angling for growth and greater stewardship/control of the brand, is a bet that you would be a better force multiplier for the movement than the next candidate. We’re now expected to believe that a leader could fail the test described here and still somehow out-strategize or out-work the next best person.
Kudos to Frances for her moral courage in the face of significant obstacles. And kudos to all the org leaders with way less experience and fewer resources who are sweating out the development of their culture, HR processes, and accountability systems. It feels like invisible work but I see it, and my advisees see it.
With this upgrade, I feel significantly more comfortable referring the site to professionals getting started in AI safety. Great work!
When hiring teams delay decisions for weeks or reject without feedback, they aren’t just reducing their chances of hiring their #1 – they’re increasing the likelihood that their #5, #6, or #7 give up on working in the space at all.
I’ve advised ~hundreds of jobseekers trying to enter similar roles, and can vouch for this author as being both highly capable and, unfortunately, dead on. This is a collective cost paid by both the candidates and the ecosystem, which 1) loses bright people and 2) takes a hit to the broader reputation.
Under very difficult constraints – which many hiring teams are – it still may be that delaying or denying feedback is the right call. But it’s costly, and I’m sorry @AnotherEAJobSeeker that you’ve been put in this situation multiple times.
You’re asking us to believe that Caroline can not only out-contribute the next best available candidate, but out-contribute that PLUS the negative externalities she brings to the project and everything associated with it. I urge you to set aside your personal feelings and consider how many projects this affects.