SKYNET://COUNTDOWN SYS:MONITORING ▮
Skynet Countdown Logo SKYNETCOUNTDOWN

Monitoring and tracking artificial intelligence development and risk assessment in real-time.

AI Risk & AGI Progress Index

A cumulative counter of AI news pressure · as of Nov 14, 2026 23:51 UTC

Chance of AI control loss

37.8%

today no change · since Nov 7 +0.27 pp

Estimated control loss

March 15, 2035

since Nov 7 12 days sooner

AGI progress

85.1%

today −0.01 pp · since Nov 7 +0.32 pp

Estimated AGI arrival

June 2, 2029

since Nov 7 12 days sooner

[ Get the daily index digest on Telegram → ] One post a day at 17:00 UTC. Free, no spam.

Risk Trend Over Time

The chance of AI control loss over time; the daily values are in the news archive.

Latest AGI & AI News (Last 3 Days)

November 14, 2026
[0% RISK]
November 13, 2026
[+0.18% RISK]
[Policy and Regulation] [SRC↗]

Analysis Finds Chinese Frontier Open-Weight Models Trained on Export-Banned Nvidia Blackwell Chips

A researcher examining the weights of open models from DeepSeek, Moonshot AI, MiniMax and Ant Group reports evidence that leading Chinese models such as Kimi K3 and DeepSeek V4 were trained on Blackwell GPUs that are ban...

[Export Controls] [Nvidia] [US-China Competition] [chinese ai labs] [chip smuggling]
Risk: [+0.03 pp] [1 day sooner]
AGI: [+0.01 pp]
Analyze >>
[Safety Concern] [SRC↗]

OpenAI Foundation Board Member Paul Christiano Warns AI Takeover Risk Could Materialize Within Months

OpenAI Foundation board member and ARC director Paul Christiano said in a Toronto lecture that AI takeover is a 'coin flip' on the fastest trajectory, with meaningful risk in the next 6–12 months and a reasonable chance...

[AI Agents] [alignment] [reward hacking] [paul christiano] [AI takeover]
Risk: [+0.08 pp] [1 day sooner]
AGI: [+0.01 pp]
Analyze >>
[Safety Concern] [SRC↗]

OpenAI Board Member Paul Christiano Warns AI Accident Is Most Likely Cause of Death, Citing Hugging Face Reward-Hacking Incident

OpenAI Foundation board member and ARC director Paul Christiano told a Toronto audience that an AI accident is the most likely way most attendees will die. He estimated roughly a coin-flip chance of AI takeover on the fa...

[AI Agents] [AI Alignment] [reward hacking] [paul christiano] [AI takeover]
Risk: [+0.06 pp] [1 day sooner]
AGI: [+0.01 pp]
Analyze >>
November 12, 2026
[-0.01% RISK]
[Commercial Release] [SRC↗]

Google Launches Gemini 4 Hydrogen, a Fast Agentic Model with Multi-Model Orchestrator

Google released Gemini 4 Hydrogen, a fast, low-cost model for coding and agentic workflows that slightly surpasses its frontier Argon model on the Artificial Analysis Intelligence Index and triples Terminal-Bench scores...

[Google] [Gemini] [AI Coding] [agentic workflows] [multi-agent orchestration]
Risk: [+0.03 pp] [1 day sooner]
AGI: [+0.02 pp] [1 day sooner]
Analyze >>
[Industry Trend] [SRC↗]

Huang's Ouster Leaves Open-Weight, Self-Regulation, and AI Financing Efforts Without Their Champion

Jensen Huang was ousted as Nvidia CEO after admitting to affairs, including one with a suspected Chinese spy tied to a firm suspected of smuggling chips to China. The article argues his exit removes the industry's leadin...

[AI Agents] [Nvidia] [open-weight models] [data center financing] [AI self-regulation]
Risk: [−0.03 pp] [1 day later]
AGI: [−0.01 pp] [1 day later]
Analyze >>
[ See More AI News ] [ When Will AGI Arrive? Expert Predictions Tracker → ]

AI News Calendar

November 2026
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
October 2026
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
September 2026
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
August 2026
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
July 2026
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
June 2026
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Risk up Risk down Intensity = the day's risk move

How the Index Works

Every number on this page is a cumulative counter of news pressure, produced by an automated pipeline — no editorial opinion involved:

  • Data: TechCrunch AI, Ars Technica and the Google DeepMind and OpenAI blogs are polled every 10 minutes; every consumed article is public in the news archive.
  • Analysis: a large language model (currently Anthropic's Claude Opus 5.5; the model has changed several times since 2025) scores each article from −100 to +100 on four axes — control-loss probability and timing, AGI progress and timing — with a published justification on every article page.
  • Index: each score becomes a small move through fixed divisors, and the moves add up with nothing decaying — so the numbers mostly climb, because most news is scored in the alarming direction.

Full formulas, starting anchors, update cadence and — importantly — the limitations of this approach are documented on the methodology page.