Search: “safety”

It’s been almost six weeks since Jensen Huang claimed we had achieved AGI. Why don’t I feel any different? https://www.forbes.com/sites/anto…

It’s been almost six weeks since Jensen Huang claimed we had achieved AGI. Why don’t I feel any different? https://www.forbes.com/sites/antoniopequenoiv/2026/03/23/nvidias-jensen-huang-says-he-thinks-

Safety

Joseph Gordon-Levitt says “almost all” AI systems are “built on mass theft,” arguing that companies using large language models “shouldn’t b…

DGX agent

Joseph Gordon-Levitt argues that most AI systems are fundamentally built on 'mass theft,' contending that companies deploying large language models without proper authorization or compensation for tra

safetygary-marcus--x

3 May 2026

Old days: “Bigshot chief scientist of major corporation can’t handle criticism of the work he hypes” New days: “Same bigshot no longer hypes…

Old days: “Bigshot chief scientist of major corporation can’t handle criticism of the work he hypes” New days: “Same bigshot no longer hypes LLMs; can’t handle the fact that someone else noticed flaws

Thank god, indeed, that @ylecun publicly stood up and admitted he was wrong, to fight for the greater good. Oh wait.

Thank god, indeed, that @ylecun publicly stood up and admitted he was wrong, to fight for the greater good. Oh wait. @GaryMarcus thank god he publicly stood up, admitted he was wrong and you were righ

that’s right.

that’s right. 😂 Finally, the great LeCun conversion arc is complete! “LLM agents = disaster” – said the guy who once defended Galactica like it was the Second Coming. Gary, you didn’t change… the time

Why is the AI backlash growing? Outside of coding (where there is clear value), and a handful of other domains (e.g. brainstorming), Generat…

Why is the AI backlash growing? Outside of coding (where there is clear value), and a handful of other domains (e.g. brainstorming), Generative AI has been a net negative for society. GenAI has been u

5 months later, I would say that AI is definitely on track to produce 90% of all bullshit within a few years. Congratulations LLM Industry, …

5 months later, I would say that AI is definitely on track to produce 90% of all bullshit within a few years. Congratulations LLM Industry, you are changing the world! Claim AI will generate 90% of kn

Are AIs about to subjugate humanity? In the debate about catastrophic AI risk, evolutionary scenarios receive far too little attention compa…

Are AIs about to subjugate humanity? In the debate about catastrophic AI risk, evolutionary scenarios receive far too little attention compared to largely speculative arguments about “instrumental con

safetydan-hendrycks--x

Dawkins earned this; Dennett did not.

This post likely refers to an academic honor or recognition that Richard Dawkins received but Daniel Dennett did not, posted as a commentary by cognitive scientist Gary Marcus on X (formerly Twitter).

Fact: I could be rich if I posted a lot of bullshit about AI.

Gary Marcus comments on financial incentives in AI discourse, suggesting that posting sensationalized or false claims about artificial intelligence could generate significant revenue. This reflects Ma

If only Richard Dawkins had seen this.

If only Richard Dawkins had seen this. 1/2 Why AI is unlikely to become conscious – my 2026 @TEDTalks is now online. What do you think about the prospects for 'conscious AI'? https://www.ted.com/talks

If you read Dawkins on AI and consciouness I hope you will take the time to also watch @anilkseth’s lucid TED talk on the same topic (link b…

If you read Dawkins on AI and consciouness I hope you will take the time to also watch @anilkseth’s lucid TED talk on the same topic (link below). @RichardDawkins Hi @RichardDawkins - the illusion of

Indeed, it does seem strange. Far more skepticism seems to have been applied to the first idea than the second.

Indeed, it does seem strange. Far more skepticism seems to have been applied to the first idea than the second. @garydanteleigh @GaryMarcus I just can't wrap my head around the fact that this guy is s

omg, conscious AI, something BIG is happening, blah blah blah

Gary Marcus, a prominent AI researcher and skeptic, posted commentary on claims about conscious AI systems. The post likely discusses concerns about overstated assertions regarding machine consciousne

People are really enjoying our full workshops showing end to end walkthroughs of real production workflows! This is a rare double header wit…

People are really enjoying our full workshops showing end to end walkthroughs of real production workflows! This is a rare double header with @braintrust's Giran Moodley and @OussamaHaff walking thoug

safetyswyx--x

suno can spit bars 😵‍💫 https://open.spotify.com/track/2ZqWAGhGt13WAt4krXv8bV?si=c953d55900a8455f ♫ Bias Alignment I’m deriving the varianc…

suno can spit bars 😵‍💫 https://open.spotify.com/track/2ZqWAGhGt13WAt4krXv8bV?si=c953d55900a8455f ♫ Bias Alignment I’m deriving the variance, bias alignment is delicate Scaling parameters, balancing gr

safetyyohei-nakajima--x

The Simpsons, on the trillion dollar buildout for GenAI.

Gary Marcus discusses concerns about the massive financial investment being directed toward generative AI development and deployment, likely examining whether the trillion-dollar spending is justified

writing a substack dissecting Richard Dawkins’ latest was not how I had planned to spend my Saturday morning 😢

Gary Marcus expressed frustration on X about unexpectedly spending his Saturday morning writing a Substack article critiquing Richard Dawkins' recent work or statements. The post suggests Marcus felt

A High-Throughput Compute-Efficient POMDP Hide-And-Seek-Engine (HASE) for Multi-Agent Operations

arXiv:2604.27162v1 Announce Type: cross Abstract: Reinforcement Learning (RL) algorithms exhibit high sample complexity, particularly when applied to Decentralized Partially Observable Markov Decision

A Woman with a Knife or A Knife with a Woman? Measuring Directional Bias Amplification in Image Captions

arXiv:2503.07878v5 Announce Type: replace-cross Abstract: When we train models on biased datasets, they not only reproduce data biases, but can worsen them at test time - a phenomenon called bias ampl

Accelerating Policy Synthesis in Large-Scale MDPs via Hierarchical Adaptive Refinement

arXiv:2506.17792v2 Announce Type: replace Abstract: Software-intensive systems, such as software product lines and robotics, utilise Markov decision processes (MDPs) to capture uncertainty and analyse

Addressing the Reality Gap: A Three-Tension Framework for Agentic AI Adoption

arXiv:2604.27245v1 Announce Type: cross Abstract: Generative AI has rapidly entered education through free consumer tools, outpacing the ability of schools and universities to respond. Now a new wave

AEGIS: Authentic Edge Growth In Sparsity for Link Prediction in Edge-Sparse Bipartite Knowledge Graphs

arXiv:2509.22017v4 Announce Type: replace Abstract: Bipartite knowledge graphs in niche domains are typically data-poor and edge-sparse, which hinders link prediction. We introduce AEGIS (Authentic Ed

Agent-Agnostic Evaluation of SQL Accuracy in Production Text-to-SQL Systems

arXiv:2604.28049v1 Announce Type: new Abstract: Text-to-SQL (T2SQL) evaluation in production environments poses fundamental challenges that existing benchmarks do not address. Current evaluation metho

Agent Name Service (ANS): A Proof-of-Concept Trust Layer for Secure AI Agent Discovery, Identity, and Governance in Kubernetes

arXiv:2604.26997v1 Announce Type: cross Abstract: Autonomous AI agent ecosystems require stronger mechanisms for secure discovery, identity verification, capability attestation, and policy governance.

Agentic Memory: Learning Unified Long-Term and Short-Term Memory Management for Large Language Model Agents

arXiv:2601.01885v2 Announce Type: replace Abstract: Large language model (LLM) agents face fundamental limitations in long-horizon reasoning due to finite context windows, making effective memory mana

safetyarxiv-cs-cl

AI and GDP - what does it mean? Sure, AI is “contributing” to GDP — but @davidsacks really nails it below, “Paying people to dig holes and f…

AI and GDP - what does it mean? Sure, AI is “contributing” to GDP — but @davidsacks really nails it below, “Paying people to dig holes and fill them back up increases GDP, as it is currently measured,

AI Models for Depressive Disorder Detection and Diagnosis: A Review

arXiv:2508.12022v2 Announce Type: replace Abstract: Major Depressive Disorder is one of the leading causes of disability worldwide, yet its diagnosis still depends largely on subjective clinical asses

AID: Agent Intent from Diffusion for Multi-Agent Informative Path Planning

arXiv:2512.02535v2 Announce Type: replace Abstract: Information gathering in large-scale or time-critical scenarios (e.g., environmental monitoring, search and rescue) requires broad coverage within l

safetyarxiv-cs-ro

An adaptive wavelet-based PINN for problems with localized high-magnitude source

arXiv:2604.28180v1 Announce Type: new Abstract: In recent years, physics-informed neural networks (PINNs) have gained significant attention for solving differential equations, although they suffer fro

Analytical Correction for Subsampling Bias in Drifting Models

arXiv:2604.27239v1 Announce Type: new Abstract: Drifting models are capable one-step generative models trained to follow a drifting field. The field combines attractive and repulsive softmax-weighted

ANCORA: Learning to Question via Manifold-Anchored Self-Play for Verifiable Reasoning

arXiv:2604.27644v1 Announce Type: cross Abstract: We propose a paradigm shift from learning to answer to learning to question: can a language model generate verifiable problems, solve them, and turn t

AttriBE: Quantifying Attribute Expressivity in Body Embeddings for Recognition and Identification

arXiv:2604.27218v1 Announce Type: new Abstract: Person re-identification (ReID) systems that match individuals across images or video frames are essential in many real-world applications. However, exi

safetyarxiv-cs-cv

Automatic Causal Fairness Analysis with LLM-Generated Reporting

arXiv:2604.27011v1 Announce Type: cross Abstract: AutoML, intended as the process of automating the application of machine learning to real-world problems, is a key step for AI popularisation. Most Au

Bayesian policy gradient and actor-critic algorithms

arXiv:2604.27563v1 Announce Type: new Abstract: Policy gradient methods are reinforcement learning algorithms that adapt a parameterized policy by following a performance gradient estimate. Convention

Belief-Guided Inference Control for Large Language Model Services via Verifiable Observations

arXiv:2604.27536v1 Announce Type: new Abstract: In black-box large language model (LLM) services, response reliability is often only partially observable at decision time, while stronger inference pat

Beyond Pixel Fidelity: Minimizing Perceptual Distortion and Color Bias in Night Photography Rendering

arXiv:2604.28136v1 Announce Type: new Abstract: Night Photography Rendering (NPR) poses a significant challenge due to the extreme contrast between dark and illuminated areas in scenes, stemming from

safetyarxiv-cs-cv

BicKD: Bilateral Contrastive Knowledge Distillation

arXiv:2602.01265v2 Announce Type: replace Abstract: Knowledge distillation (KD) is a machine learning framework that transfers knowledge from a teacher model to a student model. The vanilla KD propose

Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents

arXiv:2604.27699v1 Announce Type: new Abstract: Current embodied agents are often limited to passive instruction-following or reactive need-satisfaction, lacking a stable, high-order value framework e

CLAMP: Contrastive Learning for 3D Multi-View Action-Conditioned Robotic Manipulation Pretraining

arXiv:2602.00937v2 Announce Type: replace-cross Abstract: Leveraging pre-trained 2D image representations in behavior cloning policies has achieved great success and has become a standard approach for

ClipTBP: Clip-Pair based Temporal Boundary Prediction with Boundary-Aware Learning for Moment Retrieval

arXiv:2604.27591v1 Announce Type: cross Abstract: Video moment retrieval is the task of retrieving specific segments of a video corresponding to a given text query. Recent studies have been conducted

Co-Evolving Policy Distillation

arXiv:2604.27083v1 Announce Type: new Abstract: RLVR and OPD have become standard paradigms for post-training. We provide a unified analysis of these two paradigms in consolidating multiple expert cap

CoAX: Cognitive-Oriented Attribution eXplanation User Model of Human Understanding of AI Explanations

arXiv:2604.27354v1 Announce Type: new Abstract: Explainable AI (XAI) aims to improve user understanding and decisions when using AI models. However, despite innovations in XAI, recent user evaluations

Constrained Policy Optimization with Cantelli-Bounded Value-at-Risk

arXiv:2601.22993v3 Announce Type: replace Abstract: We introduce the Value-at-Risk Constrained Policy Optimization algorithm (VaR-CPO), a sample efficient and conservative method designed to optimize

Continuous-time q-learning for mean-field control with common noise, part-I: Theoretical foundations

arXiv:2604.27372v1 Announce Type: cross Abstract: This paper investigates the continuous-time counterpart of the Q-function for entropy-regularized mean-field control (MFC) with controlled common nois

Cost-Aware Learning

arXiv:2604.28020v1 Announce Type: new Abstract: We consider the problem of Cost-Aware Learning, where sampling different component functions of a finite-sum objective incurs different costs. The objec

Cross-Subject Generalization for EEG Decoding: A Survey of Deep Learning Methods

arXiv:2604.27033v1 Announce Type: new Abstract: Deep learning for cross-subject EEG decoding is hindered by high inter-subject variability, which introduces a severe domain shift between training and

Customer service in the age of AI has been become truly horrible. Excruciatingly bad.

Customer service in the age of AI has been become truly horrible. Excruciatingly bad. Hey @DHLCanadaHelp, your customer service really and truly sucks. You claimed to try to deliver a package, but did