Validations Queue

89,667 candidate doc → node links pending adjudication. Each was auto-generated by cosine similarity ≥ 0.55 between document and prediction embeddings (bge-base-en-v1.5, 768-dim). Showing page 214 of 833, 100 rows by similarity. Adjudicating updates doc_node_links.reviewed=true with the chosen polarity, writes per-link rows to audit_log, and removes the row from this queue. Phase 4 inference will use confirmed corroborates/contradicts links as Bayesian evidence.

Pending (total)

89,667

Reviewed

Filtered

83,275

page 214 / 833

Showing on page

100

Filters

URL params control source + min-similarity threshold

Source

All (89667)arxiv (42333)codex_research_pack (134)edgar_8k (3246)fda (4195)gdelt (1218)github_release (3087)manifold (9230)polymarket (26224)

Page size

50 100 200 500

Bulk auto-confirm by similarity threshold

Confirm or label (corroborates / contradicts) all unreviewed links above a similarity threshold in one transaction. Each affected row writes a per-link audit_log entry. Capped at 1,000 links per call. Use Preview first.

Threshold:Source:Action:

Source0 selected · 0 visible

Bulk action

Sim	Doc	Source	Pred	Domain	Prior
0.65	Beating the Style Detector: Three Hours of Agentic Research on the AI-Text Arms Race	arxiv 2026-05-04	ROB_027 The 'Paperclip Maximizer' thought experiment — unaligned superintelligence consuming all planetary resources to execute a single mundane task — is being revived as a practical engineering concern as AI transitions from digital to physical domains. The ... Nick Bostrom	AI	40%
0.65	TeamUp: Semantic Project Matching and Team Formation for Learning at Scale	arxiv 2026-05-05	IND_002 Educational systems must pivot from teaching skill acquisition and execution to teaching orchestration of agentic swarms and discovery of purpose — value accrues to individuals who master AI orchestration over traditional knowledge work; rendering trad... Alex Finn	Education	50%
0.65	When Correct Isn't Usable: Improving Structured Output Reliability in Small Language Models	arxiv 2026-05-04	238_070 Small language models will be where most algorithmic breakthroughs come from (crowdsourced) Alex Wissner-Gross	AI	42%
0.65	Bridging the Embodiment Gap: Disentangled Cross-Embodiment Video Editing	arxiv 2026-05-05	IND_013 General intelligence requires systems capable of solving tasks across multidimensional continuous domains — DreamerV3 + world-model research demonstrates scalable algorithms using latent imagination outperform traditional reinforcement learning; future... Jimmy Ba	AI	79%
0.65	Feedback Motion Planning for Stochastic Nonlinear Systems with Signal Temporal Logic Specifications	arxiv 2026-05-04	CMQ_063 Quantum compute integration with AI optimization algorithms and material-science discovery could drastically accelerate algorithmic efficiencies for Intelligence Explosion — potentially pulling superintelligence timelines closer. Alex Wissner-Gross	Quantum/AI	15%
0.65	Change-Robust Online Spatial-Semantic Topological Mapping	arxiv 2026-05-04	CMQ_056 Small Language Model (SLM) optimizations and model-distillation techniques will enable localized humanoid reasoning with extreme power efficiency — embedded AI without cloud dependency. Dario Amodei	AI/Compute	19%
0.65	AoI-Aware Multi-Robot Sensing and Transport on Connected Graphs	arxiv 2026-05-04	S_ROBOTAXI_MASS_2030 Robotaxi >10% urban miles by Nov 2030	robotaxi_deployment	30%
0.65	FDA ANDA215759: LENALIDOMIDE (LENALIDOMIDE) — BIOCON PHARMA	fda 2026-04-27	COD_BIO_001 FDA finalizes or materially advances AI-for-drug-submission guidance by end 2026 Codex Research Pack	Biotech/Longevity	47%
0.65	Why Geometric Continuity Emerges in Deep Neural Networks: Residual Connections and Rotational Symmetry Breaking	arxiv 2026-05-06	229_042 Figure believes there will be a single omni-model fusing language, vision, action, memory — not multiple specialized neural nets per task. Brett Adcock	AI	40%
0.65	Reinforcement Learning for Compositional Generalization with Outcome-Level Optimization	arxiv 2026-05-06	233_021 AI learning will improve via closed-loop reinforcement learning cycle making results keep increasing. Joe Liemandt	AI	39%
0.65	Paraphrase-Induced Output-Mode Collapse: When LLMs Break Character Under Semantically Equivalent Inputs	arxiv 2026-05-06	CMQ_056 Small Language Model (SLM) optimizations and model-distillation techniques will enable localized humanoid reasoning with extreme power efficiency — embedded AI without cloud dependency. Dario Amodei	AI/Compute	19%
0.65	Material Database Agent: A Multimodal Agentic Framework for Scientific Literature Mining	arxiv 2026-05-05	AUT_002 Models excelling at highly structured mathematical benchmarks exhibit a 'unified capability substrate' enabling dominance in seemingly unrelated fields (coding, logical reasoning, scientific discovery) — the 'mathematical skeleton' of the technological... Alex Wissner-Gross	AI	59%
0.65	FDA ANDA207974: FINGOLIMOD HYDROCHLORIDE (FINGOLIMOD HYDROCHLORIDE) — ALEMBIC	fda 2026-04-24	COD_BIO_001 FDA finalizes or materially advances AI-for-drug-submission guidance by end 2026 Codex Research Pack	Biotech/Longevity	47%
0.65	facebookresearch/habitat-lab v0.1.6	github_release 2020-09-18	AUT_017 Defining 2023-2026 shift: maturation of 'Agentic AI' via heavy fine-tuning of pre-trained LLMs using advanced reinforcement learning techniques — single foundational models seamlessly handling diverse complex tasks (market analysis, software engineerin... Jimmy Ba	AI	65%
0.65	facebookresearch/xformers v0.0.24	github_release 2024-01-31	229_028 Figure will NOT license out its neural net or hardware IP to third-party form-factor builders. Brett Adcock	Robotics	69%
0.65	Will Bitcoin dip to $45,000 by December 31, 2026?(market prob: 39%)	polymarket 2025-11-24	231_051 Entire generation growing up with crypto (Ethereum, Bitcoin) as means of exchange will have near-zero switching cost. Salim Ismail	Crypto	48%
0.65	Will this market close between 60% - 70%(market prob: 61%)	manifold 2026-05-07	246_022 Claude Mythos will NOT be released publicly near-term (PolyMarket ~7%). PolyMarket	AI	45%
0.65	Verifier-Backed Hard Problem Generation for Mathematical Reasoning	arxiv 2026-05-07	IND_013 General intelligence requires systems capable of solving tasks across multidimensional continuous domains — DreamerV3 + world-model research demonstrates scalable algorithms using latent imagination outperform traditional reinforcement learning; future... Jimmy Ba	AI	79%
0.65	When No Benchmark Exists: Validating Comparative LLM Safety Scoring Without Ground-Truth Labels	arxiv 2026-05-07	AUT_002 Models excelling at highly structured mathematical benchmarks exhibit a 'unified capability substrate' enabling dominance in seemingly unrelated fields (coding, logical reasoning, scientific discovery) — the 'mathematical skeleton' of the technological... Alex Wissner-Gross	AI	59%
0.65	Is One Layer Enough? Understanding Inference Dynamics in Tabular Foundation Models	arxiv 2026-05-07	CMQ_063 Quantum compute integration with AI optimization algorithms and material-science discovery could drastically accelerate algorithmic efficiencies for Intelligence Explosion — potentially pulling superintelligence timelines closer. Alex Wissner-Gross	Quantum/AI	15%
0.65	NavOne: One-Step Global Planning for Vision-Language Navigation on Top-Down Maps	arxiv 2026-05-07	CMQ_056 Small Language Model (SLM) optimizations and model-distillation techniques will enable localized humanoid reasoning with extreme power efficiency — embedded AI without cloud dependency. Dario Amodei	AI/Compute	19%
0.65	Cumulative-Goodness Free-Riding in Forward-Forward Networks: Real, Repairable, but Not Accuracy-Dominant	arxiv 2026-05-07	AUT_002 Models excelling at highly structured mathematical benchmarks exhibit a 'unified capability substrate' enabling dominance in seemingly unrelated fields (coding, logical reasoning, scientific discovery) — the 'mathematical skeleton' of the technological... Alex Wissner-Gross	AI	59%
0.65	FunctionalAgent: Towards end-to-end on-top functional design	arxiv 2026-05-07	SEM_022 FP4 / ternary-weight architectures decouple AI capability from raw transistor density — embargoed nations maintain competitive development. Dave Blundin	AI/Architecture	65%
0.65	Beyond Accuracy: Policy Invariance as a Reliability Test for LLM Safety Judges	arxiv 2026-05-07	AI_036 Reinforcement Learning from Human Feedback (RLHF) will fail catastrophically when applied to superintelligence — because humans will be inherently incapable of evaluating the incomprehensible logic and actions of an ASI; therefore, aligning superintell... Leopold Aschenbrenner	AI	49%
0.65	Causal Reinforcement Learning for Complex Card Games: A Magic The Gathering Benchmark	arxiv 2026-05-07	ROB_027 The 'Paperclip Maximizer' thought experiment — unaligned superintelligence consuming all planetary resources to execute a single mundane task — is being revived as a practical engineering concern as AI transitions from digital to physical domains. The ... Nick Bostrom	AI	40%
0.65	Detecting AI-Generated Videos with Spiking Neural Networks	arxiv 2026-05-07	AI_024 As AI-generated video becomes indistinguishable from reality, a massive economic resurgence in the 'Face-to-Face' (F2F) economy will emerge — live-event production, in-person experiences, and physical gatherings rebound as synthetic-reality counterweig... Mark Cuban	Consumer	45%
0.65	Crafting Reversible SFT Behaviors in Large Language Models	arxiv 2026-05-07	ROB_013 Research foundations enabling seamless human-robot interaction in complex 3D physical spaces — algorithms translating human natural-language intent directly into complex robotic actuation eliminate hard-coded robotics engineering, allowing systems to g... Jimmy Ba	Robotics	62%
0.65	Online Bayesian Calibration under Gradual and Abrupt System Changes	arxiv 2026-05-07	ROB_025 Genomic prediction will be entirely revolutionized by high-throughput phenotyping — utilizing swarms of drones and physical robots to collect orders of magnitude more trait information than historically possible by human scientists; Bayesian optimizati... Peter Dannenberg	Biotech/Longevity	70%
0.65	Layer Collapse in Diffusion Language Models	arxiv 2026-05-07	SEM_022 FP4 / ternary-weight architectures decouple AI capability from raw transistor density — embargoed nations maintain competitive development. Dave Blundin	AI/Architecture	65%
0.65	Understanding diffusion models requires rethinking (again) generalization	arxiv 2026-05-07	IND_013 General intelligence requires systems capable of solving tasks across multidimensional continuous domains — DreamerV3 + world-model research demonstrates scalable algorithms using latent imagination outperform traditional reinforcement learning; future... Jimmy Ba	AI	79%
0.65	Beyond Negative Rollouts: Positive-Only Policy Optimization with Implicit Negative Gradients	arxiv 2026-05-07	229_044 Positive transfer learning will continue to emerge, meaning more diverse data makes Figure robots broadly better at many tasks. Brett Adcock	AI	35%
0.65	The Frequency Confound in Language-Model Surprisal and Metaphor Novelty	arxiv 2026-05-07	238_070 Small language models will be where most algorithmic breakthroughs come from (crowdsourced) Alex Wissner-Gross	AI	42%
0.65	Teaching Thinking Models to Reason with Tools: A Full-Pipeline Recipe for Tool-Integrated Reasoning	arxiv 2026-05-07	CMQ_063 Quantum compute integration with AI optimization algorithms and material-science discovery could drastically accelerate algorithmic efficiencies for Intelligence Explosion — potentially pulling superintelligence timelines closer. Alex Wissner-Gross	Quantum/AI	15%
0.65	Belief Memory: Agent Memory Under Partial Observability	arxiv 2026-05-07	ROB_027 The 'Paperclip Maximizer' thought experiment — unaligned superintelligence consuming all planetary resources to execute a single mundane task — is being revived as a practical engineering concern as AI transitions from digital to physical domains. The ... Nick Bostrom	AI	40%
0.65	ReaComp: Compiling LLM Reasoning into Symbolic Solvers for Efficient Program Synthesis	arxiv 2026-05-06	CMQ_042 As AI evolves from Generative to Agentic, structural computing bottlenecks shift away from GPU and heavily toward CPU and system memory. Morgan Stanley	AI/Compute	35%
0.65	Monitoring autonomous persistent surveillance missions using invariance	arxiv 2026-05-07	SPC_015 AI will continuously analyze orbital trajectories of active satellites and space debris to predict collisions with superhuman accuracy — upon identifying collision risks, AI systems will autonomously command satellite Hall-effect thrusters to adjust or... Andrej Karpathy	Space	72%
0.65	Resource-Constrained Robotic Planning in the face of Mixed Uncertainty	arxiv 2026-05-07	AUT_017 Defining 2023-2026 shift: maturation of 'Agentic AI' via heavy fine-tuning of pre-trained LLMs using advanced reinforcement learning techniques — single foundational models seamlessly handling diverse complex tasks (market analysis, software engineerin... Jimmy Ba	AI	65%
0.65	Resource-Constrained Robotic Planning in the face of Mixed Uncertainty	arxiv 2026-05-07	ROB_027 The 'Paperclip Maximizer' thought experiment — unaligned superintelligence consuming all planetary resources to execute a single mundane task — is being revived as a practical engineering concern as AI transitions from digital to physical domains. The ... Nick Bostrom	AI	40%
0.65	AI-Aided Advancements in Autonomous Underwater Vehicle Navigation	arxiv 2026-05-06	CMQ_056 Small Language Model (SLM) optimizations and model-distillation techniques will enable localized humanoid reasoning with extreme power efficiency — embedded AI without cloud dependency. Dario Amodei	AI/Compute	19%
0.65	Will Bitcoin reach $86,000 May 4-10?(market prob: 0%)	polymarket 2026-05-04	S_IPO_TRILLION_2026 First $1T+ IPO in 2026	ipo_trillion_plus	25%
0.65	Billionaire one-time wealth tax passes in California election 2026?(market prob: 37%)	polymarket 2025-10-24	247_018 First version of social contract will be politicians writing checks in next election Dave Blundin	Macro/Economy	46%
0.65	Will Bitcoin dip to $66,000 May 11-17?(market prob: 1%)	polymarket 2026-05-11	S_IPO_TRILLION_2026 First $1T+ IPO in 2026	ipo_trillion_plus	25%
0.65	Routers Learn the Geometry of Their Experts: Geometric Coupling in Sparse Mixture-of-Experts	arxiv 2026-05-12	CMQ_063 Quantum compute integration with AI optimization algorithms and material-science discovery could drastically accelerate algorithmic efficiencies for Intelligence Explosion — potentially pulling superintelligence timelines closer. Alex Wissner-Gross	Quantum/AI	15%
0.65	Routers Learn the Geometry of Their Experts: Geometric Coupling in Sparse Mixture-of-Experts	arxiv 2026-05-12	FUT_020 Profound structural collision: exponential AI cognitive bandwidth (Webb LAMs FUT_004) colliding with Zeihan demographic atrophy (FUT_007) forces corporations + sovereign states to delegate supply chain logistics, financial forecasting, infrastructure m... Multi-Forecaster Synthesis	Labor/Jobs	45%
0.65	Agent-Based Post-Hoc Correction of Agricultural Yield Forecasts	arxiv 2026-05-12	AI_006 True autonomous agents are 'not anywhere close' — AGI and reliable long-horizon agents will require a full decade (2034 or beyond) to develop the holistic contextual reasoning and robust world models needed for unconstrained physical and digital enviro... Andrej Karpathy	AI	38%
0.65	SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces	arxiv 2026-05-12	CYB_013 The most resilient corporate moats in agentic commerce will NOT be built on software logic but on proprietary 'data flywheels' — platforms aggregating unbiased real-time consumer-intent data via continuous streams from millions of interacting autonomou... Dave Blundin	Markets/Stocks	52%
0.65	QDSB: Quantized Diffusion Schrödinger Bridges	arxiv 2026-05-12	CMQ_063 Quantum compute integration with AI optimization algorithms and material-science discovery could drastically accelerate algorithmic efficiencies for Intelligence Explosion — potentially pulling superintelligence timelines closer. Alex Wissner-Gross	Quantum/AI	15%
0.65	Understanding Sample Efficiency in Predictive Coding	arxiv 2026-05-12	CMQ_063 Quantum compute integration with AI optimization algorithms and material-science discovery could drastically accelerate algorithmic efficiencies for Intelligence Explosion — potentially pulling superintelligence timelines closer. Alex Wissner-Gross	Quantum/AI	15%
0.65	Improving the Performance and Learning Stability of Parallelizable RNNs Designed for Ultra-Low Power Applications	arxiv 2026-05-12	CMQ_042 As AI evolves from Generative to Agentic, structural computing bottlenecks shift away from GPU and heavily toward CPU and system memory. Morgan Stanley	AI/Compute	35%
0.65	A Fast and Energy-Efficient Latch-Based Memristive Analog Content-Addressable Memory	arxiv 2026-05-12	CMQ_042 As AI evolves from Generative to Agentic, structural computing bottlenecks shift away from GPU and heavily toward CPU and system memory. Morgan Stanley	AI/Compute	35%
0.65	Entropy Polarity in Reinforcement Fine-Tuning: Direction, Asymmetry, and Control	arxiv 2026-05-12	233_021 AI learning will improve via closed-loop reinforcement learning cycle making results keep increasing. Joe Liemandt	AI	39%
0.65	Is Monotonic Sampling Necessary in Diffusion Models?	arxiv 2026-05-12	235_037 Auto-regressive transformers and diffusion models will consolidate into one unified architecture. Alex Wissner-Gross	AI	35%
0.65	Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information	arxiv 2026-05-12	IND_013 General intelligence requires systems capable of solving tasks across multidimensional continuous domains — DreamerV3 + world-model research demonstrates scalable algorithms using latent imagination outperform traditional reinforcement learning; future... Jimmy Ba	AI	79%
0.65	Fast MoE Inference via Predictive Prefetching and Expert Replication	arxiv 2026-05-12	CMQ_058 Localized hardware setups (multiple Apple Mac Studios, dedicated 'AI Max 300' silicon) will allow developers to run powerful inference workloads directly on-premises — reducing cloud dependency. Alex Finn	AI/Compute	59%
0.65	Multi-Narrow Transformation as a Single-Model Ensemble: Boundary Conditions, Mechanisms, and Failure Modes	arxiv 2026-05-12	229_044 Positive transfer learning will continue to emerge, meaning more diverse data makes Figure robots broadly better at many tasks. Brett Adcock	AI	35%
0.65	Will the price of XRP be above $1.60 on May 15?(market prob: 1%)	polymarket 2026-05-08	235_041 Anthropic implied valuation using PE ratio could be one quadrillion dollars. Dave Blundin	Markets/Stocks	41%
0.65	Will the Shiller P/E Ratio hit a new all-time high before Jan 1, 2030?(market prob: 69%)	manifold 2026-05-14	CMQ_007 Anthropic corporate revenue will reach trillions of dollars before 2030. Dario Amodei	AI	41%
0.65	Safe Bayesian Optimization for Uncertain Correlations Matrices in Linear Models of Co-Regionalization	arxiv 2026-05-13	CMQ_056 Small Language Model (SLM) optimizations and model-distillation techniques will enable localized humanoid reasoning with extreme power efficiency — embedded AI without cloud dependency. Dario Amodei	AI/Compute	19%
0.65	PaMM: Periodic Motif Memory for Atomistic Models with an Explicit Local-Structure Interface	arxiv 2026-05-13	CMQ_042 As AI evolves from Generative to Agentic, structural computing bottlenecks shift away from GPU and heavily toward CPU and system memory. Morgan Stanley	AI/Compute	35%
0.65	AcquisitionSynthesis: Targeted Data Generation using Acquisition Functions	arxiv 2026-05-13	229_044 Positive transfer learning will continue to emerge, meaning more diverse data makes Figure robots broadly better at many tasks. Brett Adcock	AI	35%
0.65	Code-Centric Detection of Vulnerability-Fixing Commits: A Unified Benchmark and Empirical Study	arxiv 2026-05-13	CYB_022 Supply-chain attacks targeting widely utilized open-source AI packages — exemplified by documented attacks on the LiteLLM framework — represent extreme systemic risk: a single compromised foundational code repository instantly and autonomously infects ... Alex Wissner-Gross	AI	76%
0.65	Bridging Domain Gaps with Target-Aligned Generation for Offline Reinforcement Learning	arxiv 2026-05-13	INF_036 AI will develop a 'kinetic' edge where capital assets (drones, autonomous vehicles, agricultural robots, counter-intrusion systems) become software endpoints — requiring a bifurcated networking architecture: massive sovereign hyperscale DCs for trainin... Dara Khosrowshahi	AI	61%
0.65	Large Language Models Lack Temporal Awareness of Medical Knowledge	arxiv 2026-05-13	CYB_006 When AI agents possess the ability to read, write, and restructure their own long-term memory banks dynamically — agentically using tools to explore search spaces, understand context, recognize what's missing, and follow algorithmic curiosity — they cr... Kevin Weil	AI	63%
0.65	Widening the Gap: Exploiting LLM Quantization via Outlier Injection	arxiv 2026-05-14	SEM_022 FP4 / ternary-weight architectures decouple AI capability from raw transistor density — embargoed nations maintain competitive development. Dave Blundin	AI/Architecture	65%
0.65	An Interpretable Latency Model for Speculative Decoding in LLM Serving	arxiv 2026-05-14	238_070 Small language models will be where most algorithmic breakthroughs come from (crowdsourced) Alex Wissner-Gross	AI	42%
0.65	Octopus: History-Free Gradient Orthogonalization for Continual Learning in Multimodal Large Language Models	arxiv 2026-05-14	229_042 Figure believes there will be a single omni-model fusing language, vision, action, memory — not multiple specialized neural nets per task. Brett Adcock	AI	40%
0.65	Fast Adversarial Attacks with Gradient Prediction	arxiv 2026-05-14	SEM_047 At 200,000-GPU scale, orchestration becomes a literal 'battle against entropy' — single cosmic-ray-flipped transistor can derail 100k-GPU training run. Jimmy Ba	AI/Hardware	71%
0.65	Fast Rates for Inverse Reinforcement Learning	arxiv 2026-05-14	ROB_027 The 'Paperclip Maximizer' thought experiment — unaligned superintelligence consuming all planetary resources to execute a single mundane task — is being revived as a practical engineering concern as AI transitions from digital to physical domains. The ... Nick Bostrom	AI	40%
0.65	Scaling Laws from Sequential Feature Recovery: A Solvable Hierarchical Model	arxiv 2026-05-14	CMQ_042 As AI evolves from Generative to Agentic, structural computing bottlenecks shift away from GPU and heavily toward CPU and system memory. Morgan Stanley	AI/Compute	35%
0.65	Resolving Action Bottleneck: Agentic Reinforcement Learning Informed by Token-Level Energy	arxiv 2026-05-14	CMQ_043 In complex Agentic AI workloads, CPU-side processing accounts for 50-90% of end-to-end latency — validating CPU-bottleneck shift empirically. Morgan Stanley / Georgia Tech / Intel	AI/Compute	92%
0.65	Fully Dynamic Rebalancing in Dockless Bike-Sharing Systems via Deep Reinforcement Learning	arxiv 2026-05-14	233_021 AI learning will improve via closed-loop reinforcement learning cycle making results keep increasing. Joe Liemandt	AI	39%
0.65	Data-Augmented Game Starts for Accelerating Self-Play Exploration in Imperfect Information Games	arxiv 2026-05-14	CYB_011 Consumer commerce shifts from 'search and click' to 'ask and buy' — persistent AI agents with hyper-personalized memory navigate budgetary limits, ethical brand preferences, dietary restrictions, and logistical requirements with absolute precision, sev... Alex Finn	Consumer	44%
0.65	Price of July 2026 onion futures as of end of June?	manifold 2026-05-17	247_001 xAI will IPO at $2 trillion valuation this coming summer Peter Diamandis	AI	42%
0.65	How will reading "If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All" affect my P(Doom)?	manifold 2026-05-17	238_071 Future AI models may compress all human knowledge into megabytes via post-transformer breakthroughs Alex Wissner-Gross	AI	35%
0.65	Will the next full gemini model be as good as opus 4.7 or gpt 5.5 at coding?(market prob: 83%)	manifold 2026-05-19	238_040 Google may announce a locally hosted Gemma-type model at WWDC in June 2026 for Apple OS Alex Wissner-Gross	AI	34%
0.65	Will Perplexity become profitable by EOY 2027?(market prob: 22%)	manifold 2026-05-19	S_IPO_TRILLION_2026 First $1T+ IPO in 2026	ipo_trillion_plus	25%
0.65	Will BTC-USD close above $100,000 on Friday, June 5, 2026?(market prob: 2%)	manifold 2026-05-18	INF_068 2026 will be a watershed year for state and institutional Bitcoin accumulation — nations will drive a sustained Bitcoin shopping spree, formalizing digital assets as reserve infrastructure. Michael Saylor	Crypto	34%
0.65	Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation	arxiv 2026-05-18	CMQ_056 Small Language Model (SLM) optimizations and model-distillation techniques will enable localized humanoid reasoning with extreme power efficiency — embedded AI without cloud dependency. Dario Amodei	AI/Compute	19%
0.65	Distilling Tabular Foundation Models for Structured Health Data	arxiv 2026-05-18	CMQ_056 Small Language Model (SLM) optimizations and model-distillation techniques will enable localized humanoid reasoning with extreme power efficiency — embedded AI without cloud dependency. Dario Amodei	AI/Compute	19%
0.65	Better Together: Evaluating the Complementarity of Earth Embedding Models	arxiv 2026-05-18	SPC_016 AI transforms Earth Observation satellites via on-orbit edge computing — parsing raw visual and radar data in space, downlinking only actionable insights rather than raw imagery; permanently handing tedious low-cognitive tasks to AI and reserving human... Ben Lamm	Space	72%
0.65	Stochastic Penalty-Barrier Methods for Constrained Machine Learning	arxiv 2026-05-18	CMQ_056 Small Language Model (SLM) optimizations and model-distillation techniques will enable localized humanoid reasoning with extreme power efficiency — embedded AI without cloud dependency. Dario Amodei	AI/Compute	19%
0.65	DBES: A Systematic Benchmark and Metric Suite for Evaluating Expert Specialization in Large-Scale MoEs	arxiv 2026-05-18	CMQ_042 As AI evolves from Generative to Agentic, structural computing bottlenecks shift away from GPU and heavily toward CPU and system memory. Morgan Stanley	AI/Compute	35%
0.65	Temporal Task Diversity: Inductive Biases Under Non-Stationarity in Synthetic Sequence Modelling	arxiv 2026-05-18	AUT_002 Models excelling at highly structured mathematical benchmarks exhibit a 'unified capability substrate' enabling dominance in seemingly unrelated fields (coding, logical reasoning, scientific discovery) — the 'mathematical skeleton' of the technological... Alex Wissner-Gross	AI	59%
0.65	A Simplex Witness Certificate for Constant Collapse in Variational Autoencoders	arxiv 2026-05-18	CMQ_056 Small Language Model (SLM) optimizations and model-distillation techniques will enable localized humanoid reasoning with extreme power efficiency — embedded AI without cloud dependency. Dario Amodei	AI/Compute	19%
0.65	Dual-Rate Diffusion: Accelerating diffusion models with an interleaved heavy-light network	arxiv 2026-05-18	235_037 Auto-regressive transformers and diffusion models will consolidate into one unified architecture. Alex Wissner-Gross	AI	35%
0.65	Equilibrium Selection in Multi-Agent Policy Gradients via Opponent-Aware Basin Entry	arxiv 2026-05-18	FUT_003 Superforecaster consensus assigns 0.38% probability to AI-driven human extinction by 2100 vs domain-expert consensus of 3% — ~8x discrepancy per XPT 2022 adversarial collaboration tournament (89 superforecasters + 80 domain experts). Superforecasters m... Superforecaster Community	AI	100%
0.65	RL4RLA: Teaching ML to Discover Randomized Linear Algebra Algorithms Through Curriculum Design and Graph-Based Search	arxiv 2026-05-18	AUT_020 AI-driven adaptive learning platforms operating with near-total autonomy — human teachers transition from content deliverers to motivational 'guides' while AI handles entirety of core academic instruction. Short highly-focused 'Pomodoro' sessions manag... MacKenzie Price	Education	49%
0.65	Lightweight Gaussian Process Inference in C++ on Metal and CUDA	arxiv 2026-05-18	CMQ_058 Localized hardware setups (multiple Apple Mac Studios, dedicated 'AI Max 300' silicon) will allow developers to run powerful inference workloads directly on-premises — reducing cloud dependency. Alex Finn	AI/Compute	59%
0.65	$\boldsymbol{f}$-OPD: Stabilizing Long-Horizon On-Policy Distillation with Freshness-Aware Control	arxiv 2026-05-18	TK01 AGI Capability Plateau (2026-27 Training Stall)	—	15%
0.65	Generating Pretraining Tokens from Organic Data for Data-Bound Scaling	arxiv 2026-05-18	AUT_017 Defining 2023-2026 shift: maturation of 'Agentic AI' via heavy fine-tuning of pre-trained LLMs using advanced reinforcement learning techniques — single foundational models seamlessly handling diverse complex tasks (market analysis, software engineerin... Jimmy Ba	AI	65%
0.65	GenTS: A Comprehensive Benchmark Library for Generative Time Series Models	arxiv 2026-05-18	CYB_024 The combination of generative AI and advanced synthetic biology is actively creating a 'ChatGPT moment' for genetic engineering — historical barriers to designing advanced, highly lethal pathogens drop precipitously; necessitating new national-security... Eric Schmidt	Defense	17%
0.65	Will Bitcoin dip to $72,000 May 25-31?(market prob: 5%)	polymarket 2026-05-25	INF_069 Bitcoin's total addressable market will ultimately reach approximately $270 trillion — positioning it as "digital capital" and apex reserve asset, capturing global store-of-value demand currently distributed across real estate, gold, sovereign debt, an... Michael Saylor	Crypto	15%
0.65	Will a prediction market-related court case reach the Supreme Court before 2029?(market prob: 65%)	manifold 2026-05-27	248_010 AI capabilities will advance 100x in 2026 (by summer, potentially more). Dave Blundin	AI	31%
0.65	Double digit USD inflation before 2028 election?(market prob: 37%)	manifold 2026-05-26	236_034 Rapid inflation in near-term from UBI money printing, then rapid deflation from AI/robotics Peter Diamandis	Macro/Economy	50%
0.65	which will happen first? (Codeforces Rating)	manifold 2026-05-24	238_006 Coders will go away in the next five years (original prediction made ~3 years ago; confirmed happening in 3 years) Emad Mostaque	Labor/Jobs	40%
0.65	PhyGenHOI: Physically-Aware 4D Generation of Dynamic Human-Object Interactions	arxiv 2026-05-28	IND_013 General intelligence requires systems capable of solving tasks across multidimensional continuous domains — DreamerV3 + world-model research demonstrates scalable algorithms using latent imagination outperform traditional reinforcement learning; future... Jimmy Ba	AI	79%
0.65	PhyGenHOI: Physically-Aware 4D Generation of Dynamic Human-Object Interactions	arxiv 2026-05-28	ROB_027 The 'Paperclip Maximizer' thought experiment — unaligned superintelligence consuming all planetary resources to execute a single mundane task — is being revived as a practical engineering concern as AI transitions from digital to physical domains. The ... Nick Bostrom	AI	40%
0.65	Reinforcement Learning with Robust Rubric Rewards	arxiv 2026-05-28	ROB_027 The 'Paperclip Maximizer' thought experiment — unaligned superintelligence consuming all planetary resources to execute a single mundane task — is being revived as a practical engineering concern as AI transitions from digital to physical domains. The ... Nick Bostrom	AI	40%
0.65	When Should Models Change Their Minds? Contextual Belief Management in Large Language Models	arxiv 2026-05-28	AUT_017 Defining 2023-2026 shift: maturation of 'Agentic AI' via heavy fine-tuning of pre-trained LLMs using advanced reinforcement learning techniques — single foundational models seamlessly handling diverse complex tasks (market analysis, software engineerin... Jimmy Ba	AI	65%
0.65	Label Over Logic? How Source Cues Bias Human Fallacy Judgments More Than LLMs	arxiv 2026-05-28	248_049 Humans (weaker intelligences) can successfully align/contain super-intelligences via weak-to-strong supervision. Alex Wissner-Gross	AI	42%