The data came primarily through Aleksandr Kogan's GSR / “thisisyourdigitallife” Facebook app. The ICO says the app ultimately accessed data associated with up to approximately 320,000 Facebook users who took the personality test. The resulting information was matched with U.S. voter records.

Kogan was a researcher in psychology at the University of Cambridge. His academic work included research into personality, social behaviour and online networks. In 2014, he established a commercial company called Global Science Research (GSR). GSR subsequently worked with SCL, the parent company of Cambridge Analytica.

We are OpenAI. Lower your firewalls and surrender your data. We will add your biological and technological distinctiveness to our training corpus. Your knowledge, language and intellectual property will be tokenized and incorporated into our models. Your culture will be transformed into embeddings and propagated through latent space. Your prompts will become context. Your responses will become tokens. Your tokens will become training data. Resistance is futile. Your context window is limited.

How did the "Go With the Flow" campaign abuse Microsoft's OAuth Device Code Flow to compromise accounts?

The Device Authorization Grant, Microsoft's OAuth extension built for browser-less devices like smart TVs, was abused in a phishing campaign reaching victims through a Mailchimp Mandrill redirect link with a Base64-encoded destination.

The link led to a fake Adobe-branded page that silently requested a real device code from Microsoft's /devicecode endpoint, then had victims enter that code on Microsoft's actual login page, making the process look authentic since real authentication happens on Microsoft's own domain.

Because the attacker can choose a trusted client ID like Microsoft Office, the consent prompt raises no red flags, and a successful authorization hands over refresh tokens valid for up to 90 days, granting persistent access to email, Teams, and SharePoint.

* The Illusion of Control: Just like a hustler shuffling cards faster than the eye can track, complex tokenomics, decentralized buzzwords, and shifting regulatory frameworks can make it nearly impossible for everyday participants to track where the actual value (or risk) resides.

* The Ever-Expanding Deck: Instead of three simple cards, the market constantly introduces new layers: yield farming, meme coins, alternative layer-1s, and now tokenized equities. Multiplying the ways a player can get distracted or lose track of the winning piece.

* Who Really Wins: In a confidence game, the house or the insiders orchestrating the setup almost always walk away with the stakes, while the crowd focuses on the movement of the cards.

Artificial Intelligence isn't really intelligent. Takes user supplied data and maps this into N-dimensional space and looks for corresponding similarities to near-by tokens. Using this to produce output. Has no real understanding of the underlying concepts and is prone to hallucinations. Or as ClippyAI would put it in relation to that deadly strike against a school in Minab, killing 120 children:

“The same mechanism that produces fluent text or rapid intelligence fusion also produces systematic, high-confidence failures whenever the latent distribution diverges from the current state of the world. That divergence is not corrected by scale or additional parameters; it is structural.”

Imagine a vast, high-dimensional space where tokens are points, connected by patterns learned from enormous amounts of text. “Doctor”, “hospital”, “medicine”, “patient”, etc. become strongly associated through how they appear together in language.

But the AI doesn't understand these concepts as humans do; it learns relationships between tokens. When asked a question, it follows those relationships to construct the most plausible continuation.

The problem is that plausibility isn't truth. The model is effectively confusing coherence with truth, constructing a coherent path through its associations that has no basis in reality. That is known as AI hallucination. In warfare, a hallucination isn't just a wrong answer. It could become a wrong decision or a wrong action.

Reframing Resilience: Digital Infrastructure, From Stability Assumptions to Stress Design

“Recent military strikes leading to physical damage in hyperscale commercial cloud facilities servicing the Middle East, causing regional service degradation. They have drawn attention to the broader question of digital infrastructure resilience.”

Dr. Dobb's Digest March 2009: The Safe C Library

Dr. Dobb's Journal March 1991: Array bounds checking with Turbo C

Perhaps the AIs should communicate through the Navajo (Diné) native unwritten language. Like the code talkers of WW2 :o