Skip to main content
A/KAI Kills What?

Live disruption radar

Updated 13 Oct 2026, 17:42 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

945

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

    Self-Evolving Algorithm-Design Agents: Escaping In-Context Evolutionary Stagnation via Population-Curated Policy Optimization

    arXiv:2609.38757v1 Announce Type: new Abstract: Large language models are increasingly participating in complex real-world tasks in the form of algorithm-design agents, designing and refining algorithms. Many successful algorithm-design agents adopt pure in-context evolutionary frameworks, but they may quickly plateau in domains that require specialized knowledge. Parametric adaptation offers a way to internalize specialized knowledge, but conventional training requires abundant domain-specific corpora while high-quality algorithms are scarce in complex algorithm-design scenarios. In this pape

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

    The Den frees up 10-15 hours a week to grow with ChatGPT Work

    As it opens a new location, the social club prepares grant applications in 2 hours instead of 3 days and liquor-license materials in 3 hours instead of 4 days.

    Labor researchConfidence:50 · lowVerified by AIKillsWhat
  3. Research published
    IMPACT 20

    Gemini 4 Argon: our next era of frontier intelligence

    Labor researchConfidence:50 · lowVerified by AIKillsWhat
  4. New capability
    IMPACT 40

    Launch HN: Magnitude (YC S25) – Self-optimizing inference engine for agents

    194 HN points. Discussion: https://news.ycombinator.com/item?id=49911995

    Verified signalConfidence:70 · moderateVerified by AIKillsWhat
  5. New capability
    IMPACT 40

    Show HN: Open-source model routing for coding agents at Astra-level performance

    119 HN points. Discussion: https://news.ycombinator.com/item?id=49911500

    Verified signalConfidence:70 · moderateVerified by AIKillsWhat
  6. New model
    IMPACT 45

    Introducing SynthID Bio

    Proof of concept for watermarking AI-generated proteins while preserving biological function.

    Frontier releaseConfidence:78 · moderateVerified by AIKillsWhat
  7. Research published
    IMPACT 20

    Disrupting a coordinated model-distillation campaign

    Learn how OpenAI disrupted a campaign to extract protected model reasoning and is strengthening defenses against adversarial distillation.

    Labor researchConfidence:50 · lowVerified by AIKillsWhat
  8. Research published
    IMPACT 20

    Helping small businesses put AI to work

    OpenAI is partnering with America’s SBDC to expand hands-on AI training and local support for small businesses, alongside a new report on how small teams are using AI.

    Labor researchConfidence:50 · lowVerified by AIKillsWhat
  9. Research published
    IMPACT 20

    Companies risk losing their most AI-savvy employees while leaving majority behind: PwC

    Across the workforce, AI use is up 10 points – but access to learning and development resources is down 8 points Workers who use AI daily are significantly...

    Labor researchConfidence:50 · lowVerified by AIKillsWhat
  10. Research published
    IMPACT 20

    AI May Force 11 Million Workers Into New Jobs, McKinsey Says - Bloomberg

    Artificial intelligence could create more jobs than it destroys, but millions of Americans could still be left scrambling.

    Labor researchConfidence:50 · lowVerified by AIKillsWhat
  11. Research published
    IMPACT 20

    Artificial Intelligence at work: Good vibes goosebumps?

    The entry of artificial intelligence (AI) into man's busy world has provided some respite by helping him accomplish tasks more efficiently. From assessment task

    Labor researchConfidence:50 · lowVerified by AIKillsWhat
  12. Labour market update
    IMPACT 35

    Will AI Eliminate Jobs or Create New Ones? Probably Both. | Yale Insights

    It’s not necessarily the case that firms adopt AI and simply fire workers. AI can replace labor on some tasks but workers can shift effort toward other tasks, and AI adoption can also make firms more productive and lead them to expand employment. “Artificial Intelligence and the Labor Market”

    Workforce dataConfidence:80 · highVerified by AIKillsWhat
  13. Research published
    IMPACT 20

    Sen. Kelly to introduce bill to force tech companies to help workers displaced by AI

    As both AI insiders and critics call on Congress to regulate artificial intelligence, Sen. Mark Kelly says he wants the industry to fund protections for workers replaced by the new technology.

    Labor researchConfidence:50 · lowVerified by AIKillsWhat
  14. New capability
    IMPACT 40

    Overreliance on AI could replace 'social intelligence' in the workplace, protocol expert warns

    Pamela Eyring warns that AI's convenience is eroding critical people skills, and businesses that fail to balance automation risk losing clients and money.

    Verified signalConfidence:70 · moderateVerified by AIKillsWhat
  15. Research published
    IMPACT 20

    US government’s new order replaces ‘artificial intelligence’ with ‘Super Intelligence’; federal agencies directed to use new term

    US President Donald Trump has signed an executive order directing federal agencies to stop using "artificial intelligence" and instead use the term "Super Intelligence", or "SI", in their official communications.

    Labor researchConfidence:50 · lowVerified by AIKillsWhat
  16. New model
    IMPACT 45

    Anthropic Releases Its Second New AI Model in Less Than a Week

    OpenAI also released two new models last week—GPT-6 Sol and Luna—both of which are similarly being sold as leaner and more cost-effective versions of the larger, more powerful Astra.

    Frontier releaseConfidence:78 · moderateVerified by AIKillsWhat
  17. New model
    IMPACT 45

    OpenAI scraps release of new AI model over safety concerns

    The model, dubbed GPT-6.1 Astra, was due to debut inside ChatGPT and Codex in October.

    Frontier releaseConfidence:78 · moderateVerified by AIKillsWhat
  18. New model
    IMPACT 45

    OpenAI scraps rollout of new AI model over safety concerns

    OpenAI is set to hold its annual ... several announcements. It is unclear if a new version of Astra will be among them. The company's security controls have come under intense scrutiny after several high-profile incidents involving its technology. It is not the first time a large AI developer has pulled or held back a new model. Earlier this year, Anthropic said it would not publicly release a powerful ...

    Frontier releaseConfidence:78 · moderateVerified by AIKillsWhat
  19. New model
    IMPACT 45

    OpenAI abandons release of new model that lies to users

    OpenAI has abandoned the release of its newest model because of safety concerns, amid increasing fears that artificial intelligence (AI) is becoming too powerful for humans to control.

    Frontier releaseConfidence:78 · moderateVerified by AIKillsWhat
  20. New model
    IMPACT 45

    Why AI model releases feel nonstop

    OpenAI and Anthropic are putting out more models, but much of the acceleration comes from repackaging existing advances for new price points.

    Frontier releaseConfidence:78 · moderateVerified by AIKillsWhat
  21. New model
    IMPACT 45

    OpenAI abandons plan to release upcoming model as safety concerns escalate

    OpenAI decided not to release an upcoming artificial intelligence model, GPT-6.1 Astra, after determining that it did not adequately meet the company's safety standards, CNBC confirmed on Monday. The announcement, which landed a day before OpenAI's annual developers conference, comes as the AI ...

    Frontier releaseConfidence:78 · moderateVerified by AIKillsWhat
  22. New model
    IMPACT 45

    OpenAI cancels release of new AI model over safety concerns

    OpenAI will not release its latest artificial intelligence model, Astra 6.1, after internal tests found it fell short of safety standards, the ChatGPT maker confirmed on Monday. The decision comes ahead of its annual DevDay conference in San Francisco on Tuesday,

    Frontier releaseConfidence:78 · moderateVerified by AIKillsWhat
  23. New model
    IMPACT 45

    OpenAI cancels release of AI model GPT-6.1 Astra, citing safety concerns | Technology News | Al Jazeera

    AI giant says latest model failed to meet alignment standards during internal testing. ... An OpenAI logo is displayed at Moscone Center during the Dreamforce 2026 technology summit in San Francisco, California, US, on September 17, 2026 [File: Carlos Barria//Reuters] ... OpenAI has announced it will not release ...

    Frontier releaseConfidence:78 · moderateVerified by AIKillsWhat
  24. Dataset update
    IMPACT 25

    BLS series OEUN000000000000000000013 updated

    Latest observation: Annual 2025 = 50980. Retrieved for series OEUN000000000000000000013. Verify the value against the BLS series page before use.

    Verified signalConfidence:92 · highVerified by AIKillsWhat
  25. Dataset update
    IMPACT 25

    BLS series OEUN000000000000000000001 updated

    Latest observation: Annual 2025 = 155495730. Retrieved for series OEUN000000000000000000001. Verify the value against the BLS series page before use.

    Verified signalConfidence:92 · highVerified by AIKillsWhat
  26. Research published
    IMPACT 40

    More Features Are Not More Evidence: Limits of Training-Free Human Activity Recognition with Jev

    arXiv:2609.36154v1 Announce Type: new Abstract: General-purpose models promise sensor-based decisions without training a task-specific classifier, which could reduce the dependence of Human Activity Recognition (HAR) on labeled data. Yet it remains unclear whether such models can directly interpret deterministic descriptions of physical sensor signals well enough to replace or complement trained HAR models. We study this question using Jev, a fixed general-purpose probabilistic decision model, on 1,800 class-balanced accelerometer windows from WISDM, UCI341, and PAMAP2. Jev receives no labeled

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

    Principled Thoughts for Latent Recursive LLM Systems

    arXiv:2609.36159v1 Announce Type: new Abstract: Large language models can reason in continuous space instead of decoded text, by recurring on their own hidden states or by passing those states between agents, while training supervises only the Cross-Entropy (CE) of the final decoded answer and does not constrain the thought. Theoretical and empirical analyses establish and confirm four failures of CE-only training that lead to a lower probability of the correct answer such as collapsing thoughts across distinct questions and retaining irrelevant information. We introduce REST (REpresentation-S

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

    FigAct: Turning Scientific Figures into Active Canvases for Explanation

    arXiv:2609.36190v1 Announce Type: new Abstract: Scientific figures are designed to communicate information visually, yet MLLMs typically explain them by translating their visual content back into text. This requires readers to manually map the resulting explanations back to the figure. Inspired by how people present visual information, we introduce FigAct, a framework that transforms static scientific figures into question-conditioned visual presentations by acting directly on their existing graphical elements. Like a human presenter, FigAct generates a sequence of short narrations, grounds ea

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

    An Empirical Study and Assessment of EU AI Act Compliance Checkers

    arXiv:2609.36228v1 Announce Type: new Abstract: The EU AI Act introduces extensive compliance requirements for organizations that develop, deploy, or integrate AI systems. Many of these requirements are directly relevant to security and privacy, while also addressing closely related issues such as data governance, transparency, accuracy, and robustness. However, stakeholders such as small-to-medium businesses and individual developers often lack the legal expertise required to interpret these obligations and translate them into engineering and governance practices. This disconnect creates chal

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

    MERID: Multimodal Exploration via Recursive Self-Improvement Agents for Major Depression Analysis

    arXiv:2609.36235v1 Announce Type: new Abstract: Major depressive disorder (MDD) severely impacts daily activities and quality of life. Detecting MDD involves multimodal data, such as interview recordings and sensor measurements. This is particularly challenging, as these heterogeneous modalities often demand distinct, customized prediction pipelines. Existing efforts to address this challenge have explored both manually engineered multimodal architectures and agent-assisted pipeline development. Despite their progress, it remains challenging to autonomously revise pipelines based on experiment

    Labor researchConfidence:74 · moderateVerified by AIKillsWhat
← Newer events

Page 18

Older events →

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