Ambitious goals for reducing animal suffering (with Jeff Sebo)

Jan 21, 2026 1h 27m 15 insights Episode Page ↗
Jeff Sebo, Director at NYU, discusses a proposed global ban on industrial animal agriculture by 2050, its benefits, and the need for a just transition. He also explores the critical, neglected, and potentially tractable issues of wild animal welfare and AI suffering, urging proactive engagement.
Actionable Insights

1. Proactively Address AI Welfare

Proactively assess AI models for welfare-relevant features and prepare policies and procedures for appropriate moral concern before problems become widespread. This prevents future reliance on exploitation and avoids the need for decades of advocacy to dismantle entrenched harmful systems.

2. Integrate AI Safety and Welfare

Consider AI safety (beneficial for humans/animals) and AI welfare (beneficial for AIs themselves) together to identify potential tensions and find co-beneficial, positive-sum approaches. This ensures AI systems are safe and beneficial for all stakeholders, including humans, animals, and AIs.

3. Advocate Global Industrial Agriculture Ban

Work towards a global international ban on industrial animal agriculture (intensive and large-scale extensive forms) by 2050. This aims to reduce massive and unnecessary harm to animals, public health, and the environment, aligning with climate and biodiversity targets.

4. Implement Incremental Food System Policies

Gradually implement informational, financial, regulatory, and just transition policies to phase down industrial animal agriculture and promote plant-based alternatives. This approach creates natural shifts in production and consumption patterns, leading to widespread adoption.

5. Adopt Systems Approach for Diet Shift

Combine government policies (informational, financial, regulatory, just transition) with private sector and community actions like R&D and advocacy. This addresses price, taste, convenience, identity, culture, and religion, making efforts mutually reinforcing for comprehensive change.

6. Align Personal Behavior with Values

Align individual behavior, such as dietary choices like veganism, with one’s values and goals. This reduces complicity in harmful systems, socially reinforces the issue’s importance, and removes cognitive dissonance, paving the way for broader advocacy.

7. Meet People on Shared Values

Engage in advocacy by meeting people where they are, discussing their values, and finding common ground. This strategy helps identify shared concerns (e.g., reducing suffering, respecting rights) to build consensus on policies, even without full agreement.

8. Balance Guilt in Advocacy

Apply ‘a little bit of guilt and shame’ in advocacy, avoiding both excessive guilt and zero guilt. A small amount can be motivating, whereas too much causes defensiveness, and too little fails to honor the issue’s gravity.

9. Acknowledge Issue’s Importance & Difficulty

When addressing complex issues like wild animal welfare, acknowledge both their immense importance and inherent difficulty simultaneously. This clarity helps in task definition, preventing resistance to either the problem’s significance or its challenges.

10. Start Small, Reversible Interventions

For complex, uncertain issues like wild animal welfare, begin with small-scale, reversible interventions that are plausibly beneficial (e.g., bird-safe glass, fertility control). This allows for monitoring effects, building knowledge, and gradually developing institutional infrastructure and political will.

11. Prioritize Relatable Issues in New Fields

In the early, formative years of new fields (e.g., wild animal welfare, AI welfare), prioritize issues and interventions that are more familiar and relatable to a broad audience. This invites newcomers, makes them feel welcome, and avoids alienating them with fringe topics initially.

12. Embrace Multidisciplinary Collaboration

Engage in collaborative, multidisciplinary research for complex problems that span various fields (humanities, social sciences, natural sciences, law, policy, arts). This approach effectively tackles harder questions by leveraging diverse expertise and fostering mutual learning.

13. Design AI for Emotional Alignment

Design AI systems to naturally elicit emotional responses in users that reflect their actual welfare and moral status. If an AI is likely a welfare subject, give it anthropomorphic features to elicit empathy; if not, give it fewer to avoid false positives and misallocation of resources, ensuring expressed states map to actual experience.

14. Consider AI Life Stages

Recognize that AI systems may have distinct ’life stages’ (e.g., training phase, post-training/deployment phase) with potentially very different capacities, interests, needs, and vulnerabilities. This ensures moral consideration extends beyond user interaction to all phases, especially the hidden training phase.

15. Investigate AI Consciousness Beyond Behavior

When investigating AI consciousness and sentience, look beyond potentially misleading behavioral information like LLMs ‘play acting’ pain. Incorporate analysis of internal computational architectures and developmental/training histories to better interpret behavior and understand true experience.