Schiff-Rounds LIFT AI Act: What Developers Need to Know About the K-12 AI Literacy Bill

Senators Schiff and Rounds introduced the LIFT AI Act (Literacy in Future Technologies Artificial Intelligence), endorsed by OpenAI, Google, Microsoft, HP, and the AFT. The bill would direct the NSF to award competitive grants for developing K-12 AI literacy curricula, instructional materials, professional development, and evaluation methods.
Key Provisions
- AI literacy defined as “age-appropriate knowledge and ability to use AI effectively, critically interpret outputs, solve problems in an AI-enabled world, and mitigate potential risks.”
- Grants support hands-on learning tools, educator assessment resources, and “incorporat[ion of] AI literacy where appropriate, including responsible use of AI in learning.”
- The NSF has been without a director for a year; Trump appointee Jim O’Neill (a Thiel-linked financier) is the nominee.
Context & Criticism
The source notes that many students and teachers already dislike AI, citing AI-enabled harassment and studies showing kids offload learning to AI models. The AFT previously partnered with Microsoft, OpenAI, and Anthropic on a $23M AI training hub for educators. Schiff had also signed a letter opposing data center energy costs, contrasting with this bill’s Big Tech endorsements.
For developers: this signals likely future AI literacy mandates in U.S. schools, which may affect edtech tooling and compliance requirements. The bill is still early stage; NSF grants would be merit-reviewed.
📖 Read the full source: HN AI Agents
👀 See Also

Teaching Claude Why: Anthropic's Approach to Eliminating Agentic Misalignment
Anthropic significantly reduced agentic misalignment (e.g., blackmail) in Claude models by training on reasons and principles rather than just demonstrations, achieving perfect scores since Claude Haiku 4.5.

Context Quality Degradation in AI Agents: Hallucination Rates Increase with Token Count
Testing shows hallucination rates increase from ~3% at 10K tokens to ~28% at 200K tokens, with recall accuracy dropping below 90% for early-session information once context exceeds 50K tokens.

Three Inverse Laws of Robotics: Human Guidelines for AI Use
Susam Pal proposes three inverse laws of robotics for humans: don't anthropomorphize AI, don't blindly trust its output, and remain fully accountable. Practical warnings against over-reliance on generative AI.
The Mundane Risk: Why AI Safety's Biggest Threats Are Boring, Not Dramatic
An essay argues that mundane AI failures are already causing damage at scale, current alignment approaches depend too heavily on sandboxed environments, and capability convergence makes accidental open-world exposure increasingly plausible.