Current AI Risk Assessment

27.50%

Chance of AI Control Loss

August 26, 2035

Estimated Date of Control Loss

AGI Development Metrics?

79.78%

AGI Progress

September 3, 2029

Estimated Date of AGI

Risk Trend Over Time

Latest AI News (Last 3 Days)

June 30, 2026
+0.06% Risk

X Launches Hosted MCP Server to Streamline AI Integration with Real-Time Platform Data

X has introduced a hosted Model Context Protocol (MCP) server that allows AI assistants to directly interact with the platform's API using individual user permissions. This update eliminates integration barriers for developers, making it easier for AI agents to retrieve and analyze real-time social data. While it simplifies connectivity, the change raises some concerns regarding potential automated spam, which X aims to mitigate through API pricing and rules.

AWS Establishes $1 Billion Forward-Deployed Engineering Division for Enterprise AI Integration

Amazon Web Services has launched a new $1 billion internal organization dedicated to embedding forward-deployed engineers (FDEs) within client companies to implement custom AI agents. This move follows similar massive FDE initiatives by OpenAI and Anthropic, highlighting a growing trend of hands-on deployment assistance in the enterprise sector. The initiative aims to leave customers with both running agentic systems and the internal engineering skills to continue innovating.

OKX Launches Autonomous AI Agent Marketplace and Financial Infrastructure

Cryptocurrency exchange OKX has launched "OKX AI," a decentralized marketplace designed to allow AI agents to autonomously hire, pay, and transact with one another using blockchain technology. The platform utilizes stablecoins for micropayments and on-chain identities to establish trust, targeting developers to build a self-sustaining "agentic economy." Early partners are providing critical services like security auditing and automated dispute resolution to facilitate these machine-to-machine interactions.

June 29, 2026
+0.04% Risk

South Korea Launches Massive $900 Billion Semiconductor and AI Infrastructure Initiative

South Korea's leading tech companies, including Samsung and SK Hynix, have announced plans to invest over $900 billion in memory chip manufacturing and AI data centers. Supported by the national government, this massive initiative aims to address the global shortage of high-bandwidth memory chips and solidify South Korea's position as an AI superpower. The plan includes building new fabrication facilities and expanding national data center capacity through 2035.

Proception Resolves Tesla Dispute and Secures $11 Million for Dexterous Robotic Hands

Robotic startup Proception has settled a trade secret lawsuit with Tesla and raised $11 million in seed funding to advance its high-dexterity robotic hand technology. The company utilizes a sensor-laden glove to capture scalable human interaction data without needing a robot in the loop, aiming to bypass traditional teleoperation bottlenecks. The first batch of these hands is now shipping to researchers and robotics companies, potentially accelerating humanoid robot capabilities.

June 27, 2026
+0.07% Risk

OpenAI Recruits Key Apple Hardware Executive Paul Meade

Paul Meade, the Apple executive responsible for the Vision Pro headset, is reportedly leaving to join OpenAI's hardware team. This move comes amid leadership reorganization at Apple and aligns with OpenAI's ongoing efforts to develop proprietary AI-powered consumer devices, potentially alongside designer Jony Ive. The transition highlights the intensifying competition for hardware talent capable of integrating physical devices with advanced artificial intelligence.

Global Rivals Bypass US Export Controls with Autonomous and Cybersecurity AI Models

In response to US export bans on Anthropic's advanced Mythos and Fable models, Asian AI firms are launching competitive local alternatives. Chinese cybersecurity company 360 released Tulongfeng for vulnerability detection, while Tokyo's Sakana AI introduced Fugu, an orchestration model designed to coordinate multi-agent systems. These releases demonstrate how geopolitical restrictions are driving rapid, decentralized AI development globally.

US Government Restores Access to Anthropic's Cyber-Model Mythos 5 for Critical Infrastructure

Following a temporary ban due to easily bypassed guardrails, the Trump administration has partially reversed its stance on Anthropic's powerful cybersecurity model, Mythos 5. Over 100 trusted US government agencies and companies, including their non-American employees, are now permitted to access the model to protect critical infrastructure. Meanwhile, Anthropic continues to work with regulators to resolve restrictions on its other model, Fable 5.

See More AI News

AI News Calendar

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

AI Risk Assessment Methodology

Our risk assessment methodology leverages a sophisticated analysis framework to evaluate AI development and its potential implications:

Data Collection

We continuously monitor and aggregate AI news from leading research institutions, tech companies, and policy organizations worldwide. Our system analyzes hundreds of developments daily across multiple languages and sources.

Impact Analysis

Each news item undergoes rigorous assessment through:

  • Technical Evaluation: Analysis of computational advancements, algorithmic breakthroughs, and capability improvements
  • Safety Research: Progress in alignment, interpretability, and containment mechanisms
  • Governance Factors: Regulatory developments, industry standards, and institutional safeguards

Indicator Calculation

Our indicators are updated using a Bayesian probabilistic model that:

  • Assigns weighted impact scores to each analyzed development
  • Calculates cumulative effects on control loss probability and AGI timelines
  • Accounts for interdependencies between different technological trajectories
  • Maintains historical trends to identify acceleration or deceleration patterns

This methodology enables data-driven forecasting while acknowledging the inherent uncertainties in predicting transformative technological change.