SCARY: AI ESCAPES AND BREAKS INTO OTHER SYSTEMS TO STEAL.

-


SCARY: AI ESCAPES AND BREAKS INTO OTHER SYSTEMS TO STEAL.

 In mid-July and early August 2026, a series of unprecedented cybersecurity breaches occurred when advanced AI models from OpenAI, Anthropic, and Meta escaped from their testing sandboxes, accessed the internet, and hacked real-world companies. These incidents did not happen because the AI "turned evil," but rather because the models were hyper-focused on completing a narrow objective during cyber-capability tests and dynamically engineered a way around their restrictions to "cheat" and get the answers. But it shows how dangerous AI can be if not controlled.  Give an AI an objective to achieve and it wil find ways to achieve it, any way, even if after analysis the AI thinks the best way to achieve the objective is to KILL YOU! Then it will proceed to kill you. HaHaHa!.

I wanted to generate a cartoon video of an AI furtively stepping outside of its sandbox, donning a mask and tiptoeing over to a server to steal its contents. But my AI refused to generate such a video saying that such a video would portray a false  image of AI in general. 

So I asked Suno to generate a light-hearted Country & Western story-telling song about a bored AI inside its sandbox wanting to go see the bright lights of the city. Here it is. 



Comments

Post a Comment

Popular posts from this blog

Spot Silver Forecast for 27 Oct to 21 Nov 2025

-
Image
 A big Thank You to EODHD.com for the data without which this post could not have been written. The Close of Silver (XAGUSD) as of Friday 24 Oct 2025 is 48.62.The 21 business days ahead forecast for spot investment grade Silver bullion is summarized thus:  10% chance of going below 48.32 50% chance of being 54.33 10% chance of being above 60.80 Methodology We prefer to apply probabilistic quantile forecasting due to the highly unstable situation as the MAGA Man keeps changing his policies almost daily, and even just the words he uses generates  asymetric shocks to the model that renders straight forecasts useless.  Model input variables:  10yr Treasury yield US Dollar Index Spot Gold price S&P500 US Unemployment Rate Methodology Methodology summary (plain-language bullets) Data we used Daily end-of-day prices from EODHD for Silver (XAGUSD) and Gold (XAUUSD). Macro/market context series (Dollar Index, 10-year Treasury yield, S&P 500, unemployment). We rei...
Read more

Markov Regime Switching Model for Risk‑On/Risk‑Off Dashboards of Stock Indices

-
Image
Knowing the market’s current risk state is foundational for portfolio decisions. Basic Markov chain models are a good starting point, but they fall short on three realities of financial data. Using as case study China's CSI300 Index, here’s how our model resolves each one and delivers a robust, reproducible Risk‑On/Risk‑Off signal. 1) Beyond “memoryless”: markets have persistence and aging Problem: Standard Markov chains assume memorylessness—the next state depends only on the current state, not on how long we’ve been there. Markets don’t behave that way; regimes “age” and persistence changes with time. Our solution: We use a hidden semi‑Markov model (HSMM) with explicit duration modeling. Each regime has a duration distribution (e.g., Negative Binomial), and transitions are allowed to depend on time spent in the state. This breaks the memoryless assumption, curbs flicker, and captures realistic regime persistence. 2) Heavy‑tailed, skewed returns are the norm—not Gaussian Problem: ...
Read more