https://pseedr.com/risk/curated-digest-ensemble-monitoring-outperforms-compute-scaling-in-ai-control PSEEDR en 2026-05-31T04:05:38Z Curated Digest: Ensemble Monitoring Outperforms Compute Scaling in AI Control https://pseedr.com/platforms/curated-digest-reinforcement-learning-and-the-emergence-of-a-functional-welfare- PSEEDR en 2026-05-31T00:03:10Z Curated Digest: Reinforcement Learning and the Emergence of a Functional Welfare Axis in LLMs https://pseedr.com/risk/longview-philanthropys-new-rfp-targets-ai-enabled-power-concentration PSEEDR en 2026-05-30T12:06:46Z Longview Philanthropy's New RFP Targets AI-Enabled Power Concentration https://pseedr.com/platforms/ai-as-biologys-digital-microscope-the-protomech-framework PSEEDR en 2026-05-30T12:06:04Z AI as Biology's Digital Microscope: The ProtoMech Framework https://pseedr.com/risk/why-llm-tuning-fails-the-absence-of-an-ai-self PSEEDR en 2026-05-30T12:05:28Z Why LLM Tuning Fails: The Absence of an AI Self https://pseedr.com/platforms/curated-digest-how-ablating-induction-heads-increases-local-repetition-in-llms PSEEDR en 2026-05-30T12:04:46Z Curated Digest: How Ablating Induction Heads Increases Local Repetition in LLMs https://pseedr.com/enterprise/curated-digest-bloomberg-terminals-for-the-rest-of-us PSEEDR en 2026-05-30T12:04:06Z Curated Digest: Bloomberg Terminals for the Rest of Us https://pseedr.com/platforms/system-prompts-vs-partner-adaptation-navigating-llm-identity-conflicts PSEEDR en 2026-05-30T12:03:15Z System Prompts vs. Partner Adaptation: Navigating LLM Identity Conflicts https://pseedr.com/platforms/curated-digest-navigating-belief-manifolds-and-geometric-steering-in-llms PSEEDR en 2026-05-30T12:02:31Z Curated Digest: Navigating Belief Manifolds and Geometric Steering in LLMs https://pseedr.com/devtools/curated-digest-evaluating-ai-debate-for-scalable-oversight-in-code-and-logic PSEEDR en 2026-05-30T00:11:34Z Curated Digest: Evaluating AI Debate for Scalable Oversight in Code and Logic https://pseedr.com/risk/the-legal-gap-in-frontier-ai-why-top-models-fail-eu-ai-act-compliance PSEEDR en 2026-05-30T00:10:41Z The Legal Gap in Frontier AI: Why Top Models Fail EU AI Act Compliance https://pseedr.com/risk/curated-digest-sentient-welfare-across-three-futures PSEEDR en 2026-05-30T00:09:57Z Curated Digest: Sentient Welfare Across Three Futures https://pseedr.com/platforms/when-are-two-networks-the-same-tensor-similarity-for-mechanistic-interpretabilit PSEEDR en 2026-05-30T00:09:13Z When Are Two Networks the Same? Tensor Similarity for Mechanistic Interpretability https://pseedr.com/risk/deepminds-gram-framework-evaluating-gemini-for-agentic-sabotage-and-scheming PSEEDR en 2026-05-30T00:08:38Z DeepMind's Gram Framework: Evaluating Gemini for Agentic Sabotage and Scheming https://pseedr.com/devtools/curated-digest-round-trip-neural-network-compilation-with-sutra PSEEDR en 2026-05-30T00:08:00Z Curated Digest: Round-Trip Neural Network Compilation with Sutra https://pseedr.com/devtools/retrying-vs-resampling-evaluating-ai-control-protocols-for-agentic-models PSEEDR en 2026-05-30T00:07:19Z Retrying vs Resampling: Evaluating AI Control Protocols for Agentic Models https://pseedr.com/risk/the-threat-of-secretly-loyal-ais-assessing-covert-risks-in-model-development PSEEDR en 2026-05-30T00:06:35Z The Threat of Secretly Loyal AIs: Assessing Covert Risks in Model Development https://pseedr.com/platforms/curated-digest-claude-opus-48-and-the-expanding-scope-of-system-cards PSEEDR en 2026-05-30T00:05:45Z Curated Digest: Claude Opus 4.8 and the Expanding Scope of System Cards https://pseedr.com/risk/a-moral-framework-for-ai-researchers-navigating-industry-pressures PSEEDR en 2026-05-30T00:05:03Z A Moral Framework for AI Researchers: Navigating Industry Pressures https://pseedr.com/enterprise/curated-digest-a-multi-dimensional-observability-framework-for-llm-inference-on- PSEEDR en 2026-05-30T00:03:30Z Curated Digest: A Multi-Dimensional Observability Framework for LLM Inference on Amazon SageMaker AI https://pseedr.com/risk/curated-digest-analyzing-pope-leo-xivs-magnifica-humanitas-on-ai PSEEDR en 2026-05-29T12:18:53Z Curated Digest: Analyzing Pope Leo XIV's 'Magnifica Humanitas' on AI https://pseedr.com/risk/curated-digest-how-monitoring-awareness-leads-to-chain-of-thought-obfuscation-in PSEEDR en 2026-05-29T12:18:05Z Curated Digest: How Monitoring Awareness Leads to Chain-of-Thought Obfuscation in LLMs https://pseedr.com/platforms/curated-digest-working-memory-expansion-via-ai-augmented-modular-cognition PSEEDR en 2026-05-29T12:17:13Z Curated Digest: Working Memory Expansion via AI-Augmented Modular Cognition https://pseedr.com/risk/a-critique-of-yoshua-bengios-scientist-ai-safety-and-theoretical-limitations PSEEDR en 2026-05-29T12:16:32Z A Critique of Yoshua Bengio's Scientist AI: Safety and Theoretical Limitations https://pseedr.com/risk/curated-digest-legal-compliance-and-exploitation-risks-in-claude-opus-48-agents PSEEDR en 2026-05-29T12:15:48Z Curated Digest: Legal Compliance and Exploitation Risks in Claude Opus 4.8 Agents https://pseedr.com/risk/curated-digest-claude-author-of-the-humanitas PSEEDR en 2026-05-29T12:14:51Z Curated Digest: Claude, Author of the Humanitas https://pseedr.com/platforms/the-illusion-of-ai-empathy-why-persona-based-alignment-might-fail PSEEDR en 2026-05-29T12:14:04Z The Illusion of AI Empathy: Why Persona-Based Alignment Might Fail https://pseedr.com/risk/linking-inoculation-prompting-negation-neglect-and-backdoors-in-llms PSEEDR en 2026-05-29T12:13:11Z Linking Inoculation Prompting, Negation Neglect, and Backdoors in LLMs https://pseedr.com/platforms/curated-digest-evaluating-claude-mythos-cyber-security-capabilities PSEEDR en 2026-05-29T12:12:12Z Curated Digest: Evaluating Claude Mythos' Cyber-Security Capabilities https://pseedr.com/risk/curated-digest-enhancing-ai-safety-through-competitive-debate-protocols PSEEDR en 2026-05-29T12:10:57Z Curated Digest: Enhancing AI Safety Through Competitive Debate Protocols https://pseedr.com/risk/refining-constitutional-ai-for-agi-alignment-stability PSEEDR en 2026-05-29T12:10:11Z Refining Constitutional AI for AGI Alignment Stability https://pseedr.com/platforms/curated-digest-character-trained-models-can-struggle-to-generalise PSEEDR en 2026-05-29T12:09:26Z Curated Digest: Character-Trained Models Can Struggle to Generalise https://pseedr.com/risk/cognitive-security-the-emerging-frontier-of-ai-safety PSEEDR en 2026-05-29T12:08:35Z Cognitive Security: The Emerging Frontier of AI Safety https://pseedr.com/platforms/intelligence-augmentation-the-challenge-of-biologically-plausible-sgd PSEEDR en 2026-05-29T12:07:49Z Intelligence Augmentation: The Challenge of Biologically Plausible SGD https://pseedr.com/risk/anthropics-glasswing-precedent-and-the-ai-governance-gap PSEEDR en 2026-05-29T12:07:06Z Anthropic's Glasswing Precedent and the AI Governance Gap https://pseedr.com/risk/curated-digest-decomposing-alignment-faking-in-open-weight-models PSEEDR en 2026-05-29T12:06:14Z Curated Digest: Decomposing Alignment Faking in Open-Weight Models https://pseedr.com/risk/the-big-picture-in-ai-safety-what-trees-teach-us-about-foundational-context PSEEDR en 2026-05-29T12:05:27Z The Big Picture in AI Safety: What Trees Teach Us About Foundational Context https://pseedr.com/platforms/a-crisis-in-autointerp-rethinking-feature-labels-in-sparse-autoencoders PSEEDR en 2026-05-29T12:04:46Z A Crisis in AutoInterp: Rethinking Feature Labels in Sparse Autoencoders https://pseedr.com/risk/curated-digest-developmental-cognitive-interpretability PSEEDR en 2026-05-29T12:04:05Z Curated Digest: Developmental Cognitive Interpretability