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

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

Current AI Risk Assessment

32.08%

Chance of AI Control Loss

Estimated Date of Control Loss

June 11, 2035

AGI Development Metrics?

AGI Progress

81.90%

Estimated Date of AGI

July 21, 2029

Risk Trend Over Time

Latest AI News (Last 3 Days)

August 25, 2026
[+0.07% RISK]
[Commercial Release] [SRC↗]

OpenAI Unveils Jalapeño Inference Chip Benchmarks, Claiming Efficiency Lead Over Nvidia Blackwell

At Hot Chips, OpenAI presented the first benchmark results for Jalapeño, its custom inference chip co-developed with Broadcom, showing higher tokens per user and throughput per kilowatt than Nvidia Blackwell on Semianaly...

[OpenAI] [Compute Scaling] [custom silicon] [Broadcom] [inference hardware]
Risk: [+0.03% ↑] [-1 days ↑]
AGI: [+0.02% ↑] [-1 days ↑]
Analyze >>
[Commercial Release] [SRC↗]

OpenAI Product Lead Says Users Are Ready for Autonomous Agents as ChatGPT Work Hits 20M Users

In an interview, OpenAI core products lead Thibault Sottiaux discusses ChatGPT Work, a platform bringing Codex-style autonomous agents to white-collar workers, which he says has reached 20 million users on plans starting...

[OpenAI] [AI Agents] [codex] [chatgpt work] [ai diffusion] [inference cost]
Risk: [+0.04% ↑] [-1 days ↑]
AGI: [+0.01% ↑] [0 days]
Analyze >>
August 24, 2026
[+0.15% RISK]
[Safety Concern] [SRC↗]

Stealth Startup's Autonomous Personal AI Agent Draws Privacy and Security Alarm

Instinct, a private-access AI personal assistant from a stealth San Francisco startup led by former Sierra researcher Noah Shinn, has impressed testers with its ability to act across email, messaging, calendar, screen, a...

[AI Agents] [Autonomy] [Privacy] [prompt injection] [personal assistants] [agent security]
Risk: [+0.06% ↑] [-1 days ↑]
AGI: [+0.02% ↑] [0 days]
Analyze >>
[Industry Trend] [SRC↗]

General Intuition Nears $6B Valuation Round to Push Spatial Foundation Model Into Robotics

General Intuition, a startup building a foundation model that trains generalized agents to navigate space and time, is in talks to raise at a $6 billion pre-money valuation from Valor Equity Partners, Point72 Ventures, S...

[Robotics] [Foundation Models] [Physical AI] [world models] [venture funding]
Risk: [+0.03% ↑] [-1 days ↑]
AGI: [+0.02% ↑] [0 days]
Analyze >>
[Commercial Release] [SRC↗]

OpenAI Pushes Agentic AI Beyond Coding With ChatGPT Work, but Mainstream Adoption Lags

TechCrunch reports on ChatGPT Work, OpenAI's $20/month agentic product built from Codex that connects LLMs to email, Slack, calendars and SaaS tools so non-engineers can delegate multistep tasks. Adoption remains thin —...

[OpenAI] [AI Agents] [Enterprise Adoption] [codex] [chatgpt work] [agentic harness]
Risk: [+0.06% ↑] [-1 days ↑]
AGI: [+0.02% ↑] [0 days]
Analyze >>
August 23, 2026
[+0.02% RISK]
[Commercial Release] [SRC↗]

Anonymous 'Ox Alpha' Coding Model Appears on OpenRouter, Sparking Attribution Guessing Game

A free stealth model called Ox Alpha launched on OpenRouter, billed as a reasoning model for coding, sustained agentic work, and production workloads, with its developer choosing to remain anonymous during the preview. S...

[Reasoning Models] [agentic coding] [OpenRouter] [stealth model] [model attribution]
Risk: [+0.02% ↑] [0 days]
AGI: [+0.01% ↑] [0 days]
Analyze >>
[ Get the daily index digest on Telegram → ] One post a day at 17:00 UTC: how the Skynet Chance and AGI Progress numbers moved, and the news that moved them. Nothing else. [ See More AI News ] [ When Will AGI Arrive? Expert Predictions Tracker → ]

AI News Calendar

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
May 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
April 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
March 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
February 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
January 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
December 2025
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
November 2025
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 2025
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 2025
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 2025
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 2025
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 2025
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
May 2025
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
April 2025
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
March 2025
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
February 2025
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
January 2025
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
Risk up Risk down Intensity = impact

How the Index Works

Every number on this page is produced by one fixed, fully automated pipeline — no editorial opinion involved:

  • Data: the TechCrunch AI feed is polled every 10 minutes; every consumed article is public in the news archive.
  • Analysis: Anthropic Claude 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: scores convert to capped daily moves through fixed divisors and accumulate over time; news pushing toward more risk or faster progress is deliberately damped twice as hard as corrections.

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