Executive summary: An exploratory, steelmanning critique argues that contemporary longtermism risks amplifying a broader cultural drift toward safetyism and centralized control, is skewed by a streetlight effect toward extinction-risk work, and—when paired with hedonic utilitarian framings—can devalue individual human agency; the author proposes a more empowerment-focused, experimentation-friendly, pluralistic longtermism that also treats stable totalitarianism and “flourishing futures” as first-class priorities.

Key points:

1. Historical context & “cultural longtermism”: Longtermism is situated within a centuries-long rise in societal risk-aversion (post-WW2 liberalism, 1970s environmentalism/anti-nuclear). This tide brings real benefits but also stagnation risks that critics plausibly attribute to over-regulation and homogenizing global governance.
2. Reconciling perceptions of power: Even if explicit longtermist budgets are small, the indirect, often unseen costs of safetyist policy—slower medical progress, blocked nuclear power, NIMBY housing constraints, tabooed research—create “invisible graveyards,” making a de facto “strict culturally-longtermist state” more feasible than analysts assume.
3. Streetlight effect inside longtermism: Because extinction risks are unusually amenable to analysis and messaging, they crowd out harder-to-measure priorities—s-risks (e.g., stable totalitarianism), institutional quality, social technology, and positive-vision “flourishing futures”—potentially causing large path-dependent misallocations.
4. Utilitarian framings and the individual: Widespread (often implicit) reliance on total hedonic utilitarianism dissolves the moral salience of unique persons into interchangeable “qualia-moments” while elevating the survival of civilization as a whole—fueling totalitarian vibes and explaining why deaths of individuals (e.g., aging) receive less emphasis than civilization-level x-risk.
5. Risk of over-centralization: If longtermist x-risk agendas unintentionally bolster global regulation and control, they may increase the probability of totalitarian lock-in—the very kind of non-extinction catastrophe that longtermism underweights because it runs through messy socio-political channels.
6. Toward a more humanistic longtermism: Prioritize empowerment, experimentation, and credible-neutral social technologies (e.g., prediction markets, algorithmic policy rules, liability schemes); invest in governance concepts that reduce politicization, expand policy VOI via pluralism (charter-city-like diversity), and explicitly target anti-totalitarian interventions (propaganda/censorship-resistance, offense-defense mapping for control-enabling tech).

This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

Executive summary: In this exploratory dialogue between Audrey Tang and Plex, they probe whether “symbiogenesis” (hyperlocal, community-first cooperation that scales up) can stably beat convergent, power-seeking consequentialism, with Plex remaining skeptical that bounded/steerable systems can survive competitive pressure without a unifying, theory-level alignment that scales to superintelligence, and Audrey arguing that practicing alignment on today’s systems, strengthening defense-dominant communities, and iterating hyperlocal “civic care” and Coherent Blended Volition (CBV) may bootstrap a viable path—while both endorse improved sensemaking, shared vocabularies, and cautious experimentation.

Key points:

1. Core crux: Can complex cooperation (“symbiogenesis”) remain stable against selection for unbounded optimizers? Plex doubts boundedness survives competitive dynamics without a top-level, enforceable norm; Audrey thinks hyperlocal alignment plus defense-dominant coordination can scale and police defectors.
2. Economic pressure vs. safety: Plex argues unbounded systems will outcompete bounded/steerable ones (profit and influence gradients), making mere norms or lip service insufficient; Audrey counters with Montreal-Protocol-style, technology-forcing governance and claims steerable systems can deliver value and thus win investment.
3. Robustness requirement: Plex maintains that before strong agentic AIs, we likely need a general alignment theory that “tiles” through self-improvement (avoids sharp left turns and Goodharted proxies); Audrey frames robustness as strategy-proof rules and bounded “Kamis” (local stewards) loyal to relationships and communities.
4. Hyperlocal morality as scaffold: Audrey claims solving morality locally (quasi-utilitarianism/care ethics, subsidiarity/Ostrom) can recurse up via “dividuals” to produce stable higher-level coherence; Audrey worries local wins may aggregate into alien global outcomes that today’s humans wouldn’t endorse.
5. Coordination + sensemaking now: Both see immediate value in aligning current recommender systems, building shared vocabularies across alignment subfields, and running safe simulations (e.g., Metta’s “Clips vs. Cogs”) to test group-dynamics claims—while noting experiments won’t replace a theory expected to scale.
6. Practical implications: Focus on defense-dominant pods, transparent dashboards with awareness of Goodhart risks, participatory “utopiography”/clustered-volition processes (Weval/Global Dialogue), and cross-ontology translation; plex recommends engaging with MIRI and similar communities, and he remains cautiously supportive of AudreyThis comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.’s approach as a pathway to buy time and improve global strategy.

This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

Executive summary: This retrospective details how EAGxNigeria 2025—Africa’s largest Effective Altruism conference to date—successfully convened nearly 300 attendees from 15+ countries to strengthen EA community growth and regional engagement, while highlighting logistical, technical, and volunteer coordination lessons for future events.

Key points:

1. Scale and impact: Held in Abuja from July 11–13, 2025, EAGxNigeria hosted 295 attendees from 15+ countries, supported by $61k in expenditures and $63k in travel grants. It achieved strong satisfaction, new Giving What We Can pledges, and an average of nearly 11 new connections per participant.
2. Strategic goals: The event aimed to deepen EA engagement across Africa by connecting local and international members, emphasizing regionally relevant cause areas, and fostering collaboration on global problems.
3. Content and participation: The program included 29 sessions, 8 meetups, 21 office hours, and a popular Opportunity Fair, with attendees rating 1-1 meetings and talks as most valuable. Cause areas spanned global health, animal welfare, AI safety, and biosecurity.
4. Community outcomes: Participants reported intentions to start EA-aligned projects and organizations, apply for grants, and make giving commitments; 18 took or planned to take the Giving What We Can pledge.
5. Operational challenges: The team faced issues with badge printing, navigation, volunteer training, and minor technical faults, leading to specific recommendations such as earlier preparation, stronger AV partnerships, and more realistic volunteer simulations.
6. Volunteer coordination: 45 volunteers supported logistics, speaker liaison, and first aid, with later improvements like daily stand-ups enhancing teamwork; 44 respondents rated the experience highly despite initial confusion.
7. Lessons learned: Early planning, clearer wayfinding, and improved tech readiness were identified as key improvements for future EAGx events, along with continued investment in local leadership and volunteer capacity across Africa.

This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

Executive summary: This exploratory post argues that as the Effective Altruism (EA) and animal advocacy movements mature, many talented people may achieve greater impact by working within influential institutions—such as corporations, governments, or academia—rather than competing for limited nonprofit roles, while emphasizing that nonprofit leadership, fundraising, and entrepreneurship remain crucial exceptions.

Key points:

1. Talent bottleneck in nonprofits: The EA and animal welfare nonprofit sectors can no longer absorb the volume of mission-aligned talent, with hundreds to thousands of applicants per role and limited funding for staff expansion.
2. Strategic value of external roles: Embedding advocates in powerful institutions can yield outsized influence, giving them access to budgets, policy levers, and credibility that nonprofits can’t easily match.
3. Cost and counterfactual advantages: Working outside nonprofits conserves movement resources (since salaries are employer-funded) and offers clearer counterfactual impact—since without an advocate, the role would likely go to someone indifferent to animals.
4. Risks and caveats: External roles carry challenges such as value drift, limited animal focus, and uncertain impact; meanwhile, leadership, fundraising, and charity entrepreneurship remain high-priority nonprofit roles that are still talent-constrained.
5. Movement-level implications: Overemphasis on nonprofit careers risks wasting talent, narrowing diversity, and fostering disillusionment; a healthier distribution across sectors could make the movement more resilient and far-reaching.
6. Personal fit and discernment: Career impact depends on individual skills, motivation, and leverage of specific roles—AAC encourages exploring diverse options through personalized advising to identify the most sustainable and impactful path.

This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

Executive summary: This personal reflection celebrates Norman Borlaug as a model of practical, results-driven altruism whose agricultural innovations averted famine for hundreds of millions, arguing that his story could serve as a compelling entry point for introducing newcomers to Effective Altruism.

Key points:

1. Norman Borlaug, an agricultural scientist and Texas A&M professor, developed high-yield, disease-resistant wheat that transformed global food security during the Green Revolution, preventing massive famine in countries like Mexico and India.
2. Borlaug’s impact stemmed not only from scientific innovation but also from his persistence in field testing, farmer collaboration, and political persuasion—showing that implementation and communication are as vital as discovery.
3. His story contrasts with stereotypes of elite or Silicon Valley–based altruists: Borlaug came from poverty, learned through experience, and was driven by empathy for hunger rather than prestige or ideology.
4. Environmental opposition and institutional reluctance in the 1980s hindered Borlaug’s later efforts in Africa, which the author cites as a lesson on the importance of truth-seeking and evidence-based policy over ideology.
5. The author suggests Borlaug’s example resonates emotionally and morally with students, potentially making him a better ambassador for EA principles than abstract arguments like ITN frameworks or thought experiments.
6. The post concludes by calling for future “Borlaugs” in other high-impact areas such as food resilience under catastrophic risk, biosecurity, or AI alignment—figures who combine scientific rigor with moral urgency and real-world implementation.

This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

Executive summary: This historical and analytical essay reexamines the 19th-century Luddites to challenge the stereotype of irrational technophobia, arguing instead that they were strategic workers seeking fair terms within industrial change—and that modern labour movements confronting AI job displacement can learn from their timing, organisation, and ultimate failure.

Key points:

1. The Luddites were not anti-technology extremists but skilled textile workers responding strategically to wage cuts, deskilling, and exploitative labour practices; machine-breaking was a form of economic negotiation, not mere vandalism.
2. Their suppression by industrialists and the British government—through military force, curfews, and executions—reveals how political context and national competition shaped the limits of labour resistance.
3. “Luddite” has since become a pejorative for technophobia, though the original movement was motivated by legitimate economic and moral grievances, not hostility to progress itself.
4. Common modern takeaways (“you can’t stop technology” and the “Luddite fallacy” that automation always creates new jobs) may not hold for AI, particularly if advanced systems replace most forms of human labour.
5. Lessons for an AI-labour movement include: act early while workers retain leverage; recognise the geopolitical stakes that may discourage protest; organise legally and collectively; and treat initial activism as groundwork for later reform, not immediate victory.
6. Constructive strategies today could involve negotiating profit-sharing, legal protections, or social safety measures, rather than futilely trying to halt AI progress—learning from the Luddites’ insight but avoiding their resort to violence.

This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

Executive summary: This two-part series examines whether large language models (LLMs) can reliably detect suicide risk and explores the legal, privacy, and liability implications of their use in mental health contexts. The pilot study finds that Gemini 2.5 Flash can approximate clinical escalation patterns under controlled conditions but fails to identify indirect suicidal ideation, highlighting critical safety gaps; the accompanying policy analysis argues that current U.S. privacy and liability frameworks—especially HIPAA—are ill-equipped to govern such AI tools, calling for new laws and oversight mechanisms.

Key points:

1. Pilot findings: Gemini 2.5 Flash demonstrated structured escalation across three suicide risk levels (non-risk, ideation, imminent risk), suggesting some alignment with clinical triage behavior, but failed to detect subtle or passive suicidal expressions—posing serious safety risks.
2. Model behavior: While the LLM consistently showed empathy and offered actionable advice, it narrowed its range of supportive strategies at higher risk levels, emphasizing safety directives over coping or psychoeducation.
3. Technical and ethical implications: The study reinforces prior research showing that transformer-based models can mirror clinical reasoning but remain unreliable without domain-specific fine-tuning, ethical oversight, and crisis-specific safeguards.
4. Legal gaps: The companion analysis argues that U.S. privacy law (HIPAA) regulates by entity rather than data use, leaving AI chatbots outside its scope; instead, the FTC and new state laws (e.g., Washington’s MHMDA, California’s SB 243) are redefining “health data” to include algorithmic inferences like suicide-risk classifications.
5. De-identification challenge: Existing methods for anonymizing health data are increasingly untenable as LLMs can re-identify individuals from linguistic or contextual “fingerprints,” undermining the current legal assumption of “very small” re-identification risk.
6. Liability and governance: Clinicians remain responsible for AI-assisted decisions, but courts may soon hold developers accountable as human oversight becomes less effective; policymakers are urged to create a unified federal privacy law, mandate algorithmic transparency, and require multidisciplinary AI governance in healthcare systems.

This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

Executive summary: This reflective essay argues that while moderate Longtermism’s moral principles are broadly acceptable, its utilitarian, rationalist framing fails to motivate real-world moral action; the author proposes that genuine concern for future generations may require appealing to emotion, solidarity, and even “irrational” heroic generosity rooted in lived human experience and faith.

Key points:

1. The author distinguishes between intellectual assent to Longtermism’s claims (that future people matter) and the emotional motivation needed to act on them, arguing that moral rationalism alone inspires only a narrow subset of people.
2. Utilitarian arguments for Longtermism falter because empathy and solidarity—key drivers of moral action—depend on lived experience, which future harms cannot provide; thus, rational appeals reach a point of diminishing returns.
3. The author criticizes Essays on Longtermism for mapping moral and cognitive obstacles (like bias and myopia) without addressing how societies might actually cultivate motivation or cultural evolution toward Longtermist ethics.
4. Future ethics cannot rely on past models of moral progress, since historical empathy (e.g. toward slaves or animals) involved visible harm, unlike the abstract suffering of future generations.
5. The essay invites Longtermism to integrate broader moral resources—potentially including religious or narrative frameworks—that celebrate self-sacrifice, belonging, and love for humanity, rather than pure calculation.
6. The author concludes that safeguarding the future may require embracing forms of “heroic,” seemingly irrational generosity akin to saintly or artistic devotion, expanding Longtermism’s moral imagination beyond utilitarian rationalism.

This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

Executive summary: A candid, personal case for (maybe) applying to Forethought’s research roles by 1 November: the org’s mission is to navigate the transition to superintelligent AI by tackling underexplored, pre-paradigmatic questions; it offers a supportive, high-disagreement, philosophy-friendly environment with solid ops and management, but it’s small, not a fit for fully blue-sky independence, and the author notes cultural/worldview gaps they want to broaden (tone: invitational and self-aware rather than hard-sell; personal reflection).

Key points:

1. Roles and incentives: Forethought is hiring Senior Research Fellows (lead your own agenda) and Research Fellows (develop views while collaborating), plus possible 3–12 month visiting fellows; referral bonus up to £10k (Senior) / £5k (Research), and the application is short with a Nov 1 deadline.
2. Why Forethought: Its niche is neglected, concept-forming work on AI macrostrategy (e.g., AI-enabled coups, pathways to flourishing, intelligence explosion dynamics, existential security tools), aiming to surface questions others miss and build clearer conceptual models.
3. Research environment: Benefits include close collaboration, fast high-context feedback, protection from distortive incentives (status games, quick-karma topics), strong operations, and hands-on management that helps convert messy ideas into publishable work and sustain motivation.
4. Tradeoffs and culture: Expect some institutional asks (feedback on drafts, org decisions), not total freedom in topic choice, and a culture of open disagreement and heavy philosophy; author thinks the team’s worldview range is too narrow and collaboration/feedback timing could improve.
5. Who might fit: No specific credentials required; prized traits are curiosity about AI, comfort with pre-paradigmatic confusion, willingness to try/ditch frames, skepticism plus openness, flexibility across abstraction levels, and clear written communication.
6. Epistemic stance: This is a personal take, not an institutional pitch; the author enjoys the team and day-to-day atmosphere, acknowledges biases, and frames Forethought as a strong option for some—but not all—researchers interested in shaping society’s path through rapid AI progress.

This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

Executive summary: An exploratory, somewhat urgent argument that AI safety is losing ground to “safety washing” (performative, low-cost nods to safety) and entrenched incentives; the author contends incremental coalition-building is unlikely to suffice and urges preparing for moments of sharp Overton-window shift—most plausibly an AI disaster (secondarily mass unemployment)—by building plans, capacity, and agility to seize those openings.

Key points:

1. Diagnosis: The Paris AI Action conference and recent policy/lab developments exemplify “safety washing,” where institutions foreground trust/brand or narrow risks while sidelining catastrophic-risk mitigation; overall, AI safety has suffered setbacks across governments, labs, and legislation.
2. Incentives & rhetoric: Many actors (politicians, labs, startups, open-source, media, etc.) find AI-safety claims inconvenient and engage in motivated reasoning; common tactics include ad hominem, misdirection, overconfident assertions, and naïve skepticism—effective because they offer low-information audiences a “fig leaf.”
3. Why incrementalism struggles: The community is “outgunned,” timelines may be short, and the offense–defense balance could favor attackers; large coalitions are slow and compromise-prone, while eked-out incremental wins are likely insufficient for the level/speed of risk.
4. Strategic proposal: Shift primary planning toward leveraging rare moments of dramatic advantage—especially an AI disaster (and, secondarily, salient unemployment shocks)—that could force decisive policy; success depends on pre-planning, capacity, bravery (saying unfashionable truths), and rapid execution.
5. Contingencies & uncertainties: ChatGPT-style “wake-ups” may no longer move opinion; capabilities could plateau (in which case pivot back to growth/credibility/movement-building); national-security framings may or may not resonate with current U.S. leadership.
6. Implication for the AI-safety community: Maintain epistemic standards but get more politically realistic; invest now in preparedness for window-shifting events (e.g., agile governance mechanisms, AISI-like capacity) while downgrading expectations for near-term broad-coalition wins.

This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.