Externalities
Unintended side effects of a company's actions that affect others not directly involved.
Externalities are like the ripple effects of a company's decisions that impact people or the environment who aren't part of the original deal.
For example, if a factory pollutes a river, that's a negative externality. The factory might save money on waste disposal, but nearby towns suffer from dirty water. In AI, this could mean an AI system's behavior causes unintended harm or benefits to society.
Externalities represent market failures where the costs or benefits of an economic activity are not borne by the parties directly involved in the transaction. Positive externalities confer uncompensated benefits on third parties, while negative externalities impose uncompensated costs.
In the context of AI, negative externalities could arise from widespread job displacement due to automation, the spread of misinformation amplified by AI, or unforeseen societal shifts caused by advanced AI capabilities. Conversely, AI could generate positive externalities through accelerated scientific discovery or improved public health.
Anthropic's Long-Term Benefit Trust highlights concerns about the broad societal impacts of advanced AI. These impacts, often unforeseen, are precisely what economists call externalities. As AI systems become more powerful, understanding and managing these external effects is becoming a critical focus.
- The Long-Term Benefit TrustAnthropic · Anthropic News ·
- Anthropic Claude
- Anthropic's Long-Term Benefit Trust is designed to manage the long-term societal impacts, or externalities, of AI development.
- [object Object]
- [object Object]
The concept of externalities originated in economics, notably discussed by Arthur Cecil Pigou in his 1920 book 'The Economics of Welfare'. He analyzed how uncompensated costs or benefits could arise from economic activities. Source ↗
Try using a language model to generate summaries of complex news articles. Notice how the summary might miss nuances or introduce slight biases (a potential negative externality if widely adopted and inaccurate). Then, try asking it to explain a scientific concept; the clarity and accuracy of the explanation could be a positive externality.
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- PromptRole-play a student struggling with math for AI tutor simulationSimulate a student's math struggles and uncertainty for AI tutor evaluation. · 5 min
- TemplateReusable workflow templates for healthcare teamsTeams can create and share templates for common workflows like documentation or prior authorizations. · 15 min
- TemplateTemplate for building an AI tool tailored to your workA structured approach to defining and building a custom AI tool for specific business needs. · 15 min
- TemplateEvals CSV prompt list for agent skillsA CSV file template for listing prompts used in evaluating agent skills. · 20 min
- HeuristicDouble coding agent task completion with docs linksProviding coding agents with a documentation link significantly increases their task completion rate. · 2 min
Questions people ask
- Are externalities always bad?
- No, externalities can be positive, like when a company invests in research that later benefits other industries, or negative, like pollution from a factory.
- How do AI companies deal with externalities?
- Companies are increasingly establishing ethical guidelines, safety protocols, and governance structures, like Anthropic's trust, to anticipate and mitigate potential negative externalities.
First seen on Subvolts Jul 9, 2026. Updated Sep 14, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.