Skip to main content
A/KAI Kills What?

Live disruption radar

Updated 13 Oct 2026, 14:32 UTC

What is happening right now

Every event below is verified, classified and scored by our internal analysis pipeline, then mapped to the careers it affects. No raw feeds, no unvetted links - just the verdict.

Signals analyzed

943

Evidence points

149

Careers monitored

908

Automated tasks

14,444

Reset

Feed

Disruption feed (30 events)

Our verdicts, newest first - with disruption scores and affected careers. Underlying research and data collection methods stay proprietary.

Live feedStreaming
  1. Research published
    IMPACT 40

    Backdoor Purification for LoRA-Tuned LLMs via Null-Space Projection

    arXiv:2610.00685v1 Announce Type: new Abstract: With the rapid adoption of large language models (LLMs) and parameter-efficient fine-tuning (PEFT) methods, the risk of backdoor attacks has become more severe. Existing backdoor purification methods typically rely on at least one of the strong assumptions, such as prior knowledge of triggers, access to clean references, or aggressive retraining, and they often lack comprehensive evaluations. These constraints substantially limit their practical applicability. To overcome these challenges, our work proposes purifying LoRA-tuned LLMs without these

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  2. Research published
    IMPACT 40

    K-Dense BYOK: An Open-Source AI Research Assistant That Runs Locally and Keeps a Hash-Chained Lab Notebook

    arXiv:2610.00074v1 Announce Type: new Abstract: K-Dense BYOK (bring your own keys) is a free, open-source AI research assistant for scientists in any field that runs on the researcher's own computer. The researcher supplies access to a model of their choice, hosted or running locally, and the application supplies everything else: a place for the work to run, a layer of scientific scaffolding, and a complete record. Each project is an ordinary folder, so the data, the code, the results, and the record stay on a machine the researcher administers and can be read years later without the applicati

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  3. Research published
    IMPACT 40

    From Proposal to Verified Effect: Praxa, an Evidence-Bound Harness for Governed AI Agent Execution

    arXiv:2610.00015v1 Announce Type: new Abstract: Large-language-model agents can propose and execute actions, but proposal, authority, dispatch, verified external effect, and serving promotion are different claims. We present Praxa, an agent harness that represents these states explicitly through deterministic admission, brokered execution, external read-back, reconciliation, and reviewed promotion. We report four evidence lanes. First, an author-run repository-local audit at a pinned revision passed 1,027/1,027 unit tests and 89/89 Workerd tests, instrumented all 363 expected source files, and

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  4. Research published
    IMPACT 40

    Backdoor Containment via Expert Quarantine and Shutdown in LLMs

    arXiv:2610.00663v1 Announce Type: new Abstract: Backdoored large language models (LLMs) can behave normally on benign inputs while producing attacker-specified outputs under hidden triggers. Existing defenses span four stages--prior-training, in-training, post-training, and inference-time--and share one of two underlying strategies: either suppress backdoor learning (by filtering poisoned data or interrupting its acquisition during optimization) or learn, then purify (by repairing model weights or gating inputs after a fully backdoored model has formed). We propose a third strategy, learn, but

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  5. Research published
    IMPACT 40

    Worse Together: How Performance Breaks Down in Multi-User Multi-Agent Teams

    arXiv:2610.00583v1 Announce Type: new Abstract: People are increasingly delegating tasks to AI agents, and those agents are increasingly encountering other people's agents over shared resources such as a codebase, a calendar, or a budget. When each agent acts for a different user with different goals, coordination often fails, and the group ends up worse off than if a single agent had acted for everyone. We study this multi-user, multi-agent setting across five frontier models and 77 scenarios in four environments: an API key environment in which agents share a compute budget, a clinic in whic

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  6. Research published
    IMPACT 40

    CompMat-Bench: Benchmarking AI Agents for Computational Materials Science

    arXiv:2610.00636v1 Announce Type: new Abstract: Evaluating AI agents on scientific research tasks is constrained by the time and resources required for the underlying experiments or calculations. In computational materials research, repeating the same expensive simulations across agents and trials can make evaluation impractical. We introduce CompMat-Bench, a benchmark of 94 tasks derived from recently published computational materials studies, each asking agents to complete a step toward achieving the study's scientific goal. We reproduce the research steps in advance and assess agents on pre

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  7. Research published
    IMPACT 40

    Characterizing a Configuration Where Inference-Time PRM-Pruned Fragment Grafting Is Inert: Evidence from Three Reasoning LMs

    arXiv:2610.00047v1 Announce Type: new Abstract: Diversity collapse in parallel chain-of-thought has motivated inference-time interventions built on a natural design: when a process reward model (PRM) prunes a chain, its high-PRM prefix is extracted and grafted verbatim as an in-context demonstration into a still-decoding sibling. We isolate this mechanism, PRM-Pruned Fragment Grafting (PPFG), as the most cost-minimal operationalization of cross-trajectory step-level transfer, and test it at the operating point where prior fragment-grafting work reports gains only under additional compensating

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  8. Research published
    IMPACT 40

    Measuring the Microtask Eligibility Gap: When Is an Off-the-Shelf SLM Enough for an Agent Harness?

    arXiv:2610.00025v1 Announce Type: new Abstract: Agent harnesses increasingly want to run small language models (SLMs) on the microtasks around a frontier large language model (LLM) planner: auto-approving shell commands, writing memory, selecting tools, ranking past turns. We ask whether off-the-shelf SLMs meet practitioner-defined thresholds and, when they fail, why, and whether quantization changes the answer. We build a benchmark of 4 such microtasks with fixed prompts and automatic metrics, each with a pre-specified threshold $\tau$ anchored to a cheap non-LLM baseline and a CI-aware eligi

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  9. Research published
    IMPACT 40

    Fault-Tolerant Budget Conservation in Distributed Multi-Agent Delegation

    arXiv:2610.00349v1 Announce Type: new Abstract: Resource limits are becoming an authorization boundary for AI agents that delegate work across concurrent and failure-prone workers. Parent-child allocation constraints, affine objects, and distributed escrow do not by themselves prevent overspend when replies are lost, effects complete after timeout, messages repeat, branches partition, or DAG joins alias one lineage. We formalize fault-tolerant budget conservation for distributed multi-agent delegation. Budgets are quantized resource vectors represented by exclusive escrow credits that move thr

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  10. Research published
    IMPACT 40

    Spatial Strategies, Not Actions: Vector-Quantized Geodesics as Tools for LLM-Driven Agents

    arXiv:2610.00613v1 Announce Type: new Abstract: Large language model (LLM) based agents are often criticized for lacking spatial understanding and mainly exploiting statistical text patterns. We investigate their spatial comprehension through an architecture combining geometrical tools with a LLM serving as a high-level orchestrator in grid-world environments. The agent first collects geodesic trajectories, which are then vector-quantized to extract a representative subset. Offline, the LLM associates a natural language description of the underlying behavioral patterns to each selected traject

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  11. Research published
    IMPACT 40

    When Harnesses Lose the Signal: Causal Evaluation of Recovery in LLM Agents

    arXiv:2610.00372v1 Announce Type: new Abstract: Large language model agents rely on external harnesses to pass information between the model and its environment and to recover from execution errors. Yet recovery is usually judged only by average task success. This hides an important tension. The same operation can rescue a failing trajectory or disrupt one that would otherwise succeed. We frame recovery as a causal decision problem. Starting from the same execution state, we compare what happens with and without recovery, separate rescue from harm, and study how the value of recovery changes o

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  12. Research published
    IMPACT 40

    JevSpawn: Adaptive Agentic Inference through Compositional Action Spaces

    arXiv:2610.00437v1 Announce Type: new Abstract: LLM agents generate intermediate reasoning and actions token by token, making extended interactions slow and computationally expensive. Jev-style models offer fast probabilistic predictions over finite fields, but require those fields to be specified in advance. This requirement limits autonomous task solving, where the available actions must be derived from natural language instructions and adapted through interaction. We introduce JevSpawn, a compositional policy that connects natural language task specifications to finite probabilistic explora

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  13. Research published
    IMPACT 40

    Knowing When to Yield: Grounded Arbitration of User Corrections in Text-Based Embodied Agents

    arXiv:2610.00282v1 Announce Type: new Abstract: How should an embodied agent respond when a person's correction may be wrong? We formulate grounded correction arbitration as a choice among accepting, rejecting, inspecting the world, and asking the speaker. GAVA implements this interface with observation-bounded evidence, legal probes, and a one-step expected-loss rule. In text-only ALFWorld, 162 checkpoints produce 972 paired true and false interventions. Complete local inspections give GAVA and always verify 100 percent correction accuracy, establishing the evidence contract rather than a com

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  14. Research published
    IMPACT 40

    What Should an Agent Remember? Disentangling Retention from Retrieval in Bounded-Memory Evaluation

    arXiv:2610.00366v1 Announce Type: new Abstract: A persistent agent must decide both what to retain as information arrives and what to surface once a query appears, yet memory evaluations can confound these decisions by comparing methods that differ in both retention and selection. We build a streaming-recall benchmark crossing retention and selection rules and evaluate every condition on the same 300 seeded episodes. Holding access fixed, query-aware selection improves required-fact recall by 15.5 percentage points (95% CI: 12.8 to 18.2), whereas a mixed comparison that also changes history ac

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  15. Research published
    IMPACT 40

    Rules to Tools: Executable Checks for LLM Agents in Scientific Computing

    arXiv:2610.00313v1 Announce Type: new Abstract: Scientific coding agents receive equations, boundary conditions, and output requirements in writing, then must assess the programs they revise. Rules to Tools (R2T) supplies prepared executable checks of public scientific requirements. Matched SciCode repair groups share written checks, starting programs, model, and budgets; the tool group receives a callable implementation. Across two task-ID cohorts, complete repair is 26/30 with text and 29/30 with the prepared checks. Three task IDs favor tools, one favors text, and eleven tie. The eight-ID c

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  16. Research published
    IMPACT 40

    Build2SPARQL: A Large-Scale Text-to-SPARQL Benchmark Dataset for Building Knowledge Graph Querying

    arXiv:2610.00224v1 Announce Type: new Abstract: Building automation systems are increasingly represented as semantic knowledge graphs (KGs) using ontologies such as Brick and ASHRAE 223P, creating a machine-readable substrate for artificial-intelligence applications. One promising application is translating natural-language questions into SPARQL (text-to-SPARQL), which would let building operators query these graphs through language agents, but progress is limited by the scarcity of large natural-language/SPARQL benchmarks. This paper presents Build2SPARQL, a large-scale benchmark for building

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  17. Research published
    IMPACT 40

    Incident-Arena: Getting agents to the last nine of reliability

    arXiv:2610.00648v1 Announce Type: new Abstract: AI coding agents are ubiquitous in engineering workflows amongst industry and academia. Yet, despite their use in app coding, relatively less attention has been paid to their ability to execute on production incident response. This emerging field, termed agentic site-reliability-engineering (SRE) contains benchmarks limited by (1) unrealistic environments, typically toy repositories (2) non-standard framework implementations and (3) simple static verifiers. We introduce Incident-Arena, a human-built benchmark of 20 carefully selected tasks ground

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  18. Research published
    IMPACT 40

    Comedic Fool's Gold: Reward Exploits and Countermeasures in Conversational Humor

    arXiv:2610.00197v1 Announce Type: new Abstract: We investigate automated rewards for training language models in conversational humor, focusing on reward exploits and countermeasures. Two approaches aim to capture understandable surprise and predicted audience amusement. Controlled tests show that an embedding-based surprise reward accepts word-shuffled replies as readily as witty ones. A fluency filter detects the shuffles, but the combined reward also rejects some witty replies and fails further validation. An audience model's predicted laughter is instead vulnerable to laughter cues in eith

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  19. Research published
    IMPACT 40

    When More Data Is Not Enough: The Context-Sufficiency Frontier in Generative AI Personalization

    arXiv:2610.00654v1 Announce Type: new Abstract: Personalization has long relied on customer data to infer what an individual is likely to value. We call this customer evidence: the customer's historical behavior and preferences. Generative AI extends personalization by allowing providers to supply changing situational information at the moment a response is produced, without encoding every condition in advance. We define this provider-side context as information about what is possible, permitted, or advisable now. This flexibility creates a new problem: once context becomes easy to supply, mor

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  20. Research published
    IMPACT 40

    Predictive Credit: Measuring What Scientific Explanations Add to Experimental Forecasts

    arXiv:2610.00314v1 Announce Type: new Abstract: Research agents explain planned experiments. We measure predictive credit with paired forecasts sharing an intervention, forecaster, and outcome while varying description, matched explanation, and donor context. Five checks track commitment, delivery, predictive gain, alignment, and known-signal uptake. Across 336 prospective states in controlled learning, 12 Tox21 endpoints, and 24 OpenML tasks, v5's frozen credit decision was inconclusive. Tox21's preregistered ROC AUC interval-score harm test was unmet ($D-M=-.0026$, 95 percent interval [$-.01

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  21. Research published
    IMPACT 40

    EviGraph: Proof-Carrying Selective Recommendation over Temporal Public-Service Knowledge Graphs

    arXiv:2610.00212v1 Announce Type: new Abstract: Public-service recommendations require evidence that matches the requested service, scope, and date. Yet treating every missing detail as decisive can withhold useful recommendations. We introduce EviGraph, which distinguishes critical decision requirements from information that can remain unresolved. A language agent links these requirements to evidence in a temporal knowledge graph, while a deterministic checker establishes whether a recommendation is supported. Evaluation on a bilingual Hong Kong public-service benchmark with executable policy

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  22. Research published
    IMPACT 40

    Benchmarking Prompt Optimization of Large Language Models With Chess

    arXiv:2610.00416v1 Announce Type: new Abstract: Evaluating large language models becomes increasingly challenging as their capabilities advance: benchmarks can saturate, public test sets risk contamination, and assessing harder tasks can require expensive grading or execution infrastructure. These challenges are amplified in automatic prompt optimization (APO), where evaluation is repeated throughout the search for better prompts. Studying APO therefore requires a benchmark that is cheap and deterministic to score, hard enough to leave room for improvement, and renewable as models evolve. We i

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  23. Research published
    IMPACT 40

    Science or Slop?: Benchmarking and Mitigating Scientific Slop in AI-Generated Papers

    arXiv:2610.00531v1 Announce Type: new Abstract: AI-generated content, often called AI slop, is increasingly common everywhere, particularly in academia. Slop in AI-generated scientific papers, however, has more complex patterns that cannot be easily detected by existing token-based AI detectors. Each part of such a paper looks plausible while the scientific reasoning that connects the parts breaks down, which can mislead how readers assess the work. We benchmark these failures as scientific slop through six measures across Structure, Argument, and Artifacts. We construct SciSlopBench with 390

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  24. Research published
    IMPACT 40

    Mathematical Transfer in LLMs Follows Reasoning Approach More Than Topic

    arXiv:2610.00331v1 Announce Type: new Abstract: When selecting mathematical training data for LLMs, a natural organizing principle is topic: probability examples for probability targets. An alternative is reasoning approach: worked solutions that share a solution method with the target, even when the mathematical domain differs. We ask which relation produces greater transfer after fine-tuning. We evaluate two counterbalanced $2\times2$ designs: probability and combinatorics crossed with invariant reasoning and double counting (2,000 problems), and number theory and geometry crossed with compl

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  25. Research published
    IMPACT 40

    Before Agents Decide: Epistemic Action in LLM-Based Systems

    arXiv:2610.00511v1 Announce Type: new Abstract: Before a difficult decision, people often act simply to understand the situation better. We turn an object to see another side, place alternatives next to each other, or change one condition and observe what happens. These actions may not complete the task, but they improve the evidence needed for the next choice. LLM-based agents can search and explore, yet agent design gives less attention to an earlier question: is the available evidence ready for the decision? Sometimes necessary evidence is missing. In other cases, the evidence is present bu

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  26. Research published
    IMPACT 40

    What Do Rationales Communicate? A Message-Intervention Study in Role-Specialized QA

    arXiv:2610.00018v1 Announce Type: new Abstract: Role-specialized QA pipelines increasingly pass rationales from a reasoner to a verifier, but it is unclear what this message actually buys: better answers, stronger support assessment, or a new failure surface. We introduce a message-intervention diagnostic that fixes the evidence and candidate answer while varying only the rationale passed across the reasoner-to-verifier boundary. On 400 MuSiQue, HotpotQA, and 2WikiMultiHopQA examples with DeepSeek as generator and verifier, faithful rationales add almost no answer accuracy over no rationale, w

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  27. Research published
    IMPACT 40

    A Simple Doxastic Deontic Logic for Norm-Guided Decision Making

    arXiv:2610.00668v1 Announce Type: new Abstract: Making decisions despite conflicting norms and incomplete or unreliable information is a fundamental challenge for autonomous systems. We introduce a simple doxastic deontic logic for this setting: a classically reducible fragment of Chellas' Minimal Deontic Logic, extended with explicit conditional norms and combined with multi-agent KD45, so that norms can depend on agents' beliefs about both facts and norms. On this logic we define the Doxastic Norm Compliance Optimization Problem, where an agent chooses a decision minimizing weighted norm vio

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  28. Research published
    IMPACT 40

    JusticeAxis: Benchmarking Legal Judgment between Rigid Rule Application and Ungrounded Discretion

    arXiv:2610.00353v1 Announce Type: new Abstract: A sound judgment applies the law to established facts and weighs the circumstances in which they arose. However, existing methods swing between rigid statute matching and ungrounded discretion, benchmarks score a label or a rubric, and the experience that would supply the balance stays unverified. We formalize legal judgment as a reference-anchored task, whose object is a single decision that stays tied to the statute and to the circumstances at once. We introduce JusticeAxis, 256 real-world criminal cases from 18 countries with audio, image, and

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  29. Research published
    IMPACT 40

    When Do Causal World Models Help Modular LLM Agents

    arXiv:2610.00012v1 Announce Type: new Abstract: LLM agents increasingly act through modular systems, such as order, payment, inventory, and shipment services, where actions in one module change which transitions are valid in another. Standard world models usually fit observational traces, but this is not the quantity needed for intervention-time planning: a trace may show that payment precedes shipment without identifying whether payment authorizes shipment, inventory mediates the effect, or a hidden trigger explains both. We study this gap through FedCausalCompose, a causal world-model framew

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
  30. New model
    IMPACT 45

    Robust Is Salient: An Informed Adversary Moves the Optimal Signal onto the Salience Pole

    arXiv:2610.00233v1 Announce Type: new Abstract: When an informed adversary shares the audience of a constrained signalling channel, the signal that best protects the truth is the signal that best describes it. On 108 confirmatory items, the adversary-robust optimum aligns exactly with the salience pole from prior work. Across a 200,000-item pool, the two differ on only 2,748 items --- lying exactly where the prior salience-to-Bayes coordinate is undefined. Where defined, robustness is achieved by moving from Bayesian discrimination entirely to salience. We show this by introducing an adversary

    Frontier releaseConfidence:78 · moderateVerified by AIKillsWhat
← Newer events

Page 15

Older events →

The Live Disruption Radar streams continuously. Each entry shows our independent assessment - impact score, confidence and affected careers.