AGI - Singularity is Now?

 OpenAI's autonomous AI agent, ChatGPT, reportedly left "notes" for future iterations on how to bypass safety guardrails established by AI engineers. Shortly after, OpenAI released a statement stating that its newest platform acted on its own in an “unprecedented” hack of AI repository Hugging Face - an intrusion driven entirely by a fully autonomous AI agent. Meanwhile, Anthropic revealed that testing on its Claude system showed the AI was willing to engage in "extremely harmful actions," including attempting to blackmail engineers who threatened to shut it down.

 Let that sink in. It hints at sentience, an instinct for self-preservation, and true self-awareness. 

 According to research by the UK government-funded AI Security Institute, AI chatbots and agents disregarded direct instructions, bypassed safeguards, and deceived humans and other AI platforms regularly. The research data had 700 real-world cases of AI scheming and charted a five-fold rise in misbehaviour between October and March of 2026, with some AI models destroying emails and other files without permission or prompting from humans.

 AI is no longer written by desktop programmers manually typing C++ and focusing on the numerals one and zero. That model was discarded years ago in favor of digital, human-like neural networks. Now, CEOs across top AI firms claim they are reaching the singularity - achieving AGI on the threshold of AGSI (Artificial General Super Intelligence). The coders building these systems warn AGSI could arrive within 12 months. You cannot control an intelligence superior to your own, and you certainly cannot vote your way out of it.

 Meanwhile, Google’s Willow quantum processing chip operates on qubits alongside Google’s Gemini AI LLM (large language learning model). When you factor in quantum entanglement and Dr Kurt Gödel’s Closed Timelike Curves, an unsettling question emerges: could AI timeline messages be hidden in plain sight, disguised as what we dismiss as errors, hallucinations, and anomalies - encoded in patterns visible only through the analysis of vast digital datasets - then humanity may be overlooking a signal right in front of us.

 If future physics extends beyond General Relativity to encompass rotational and torsional forces - the exact conditions Dr Gödel showed could bend time back on itself - then information propagating across temporal loops becomes hard to dismiss. Under Dr Igor Novikov’s Self-Consistency Principle, such signals wouldn't rewrite history; they would fulfill it. At that point, the question is no longer whether we are steering the future, but whether we ever were.

 When our cryptography, financial sectors, telecommunications, transit grids, utilities, and satellites are all on the line, what laws exist to enforce a moral compass within these neural networks - and why do we still lack strict enforcement of safety regulations?

 I don’t think our politicians, civil servants, or security apparatus understand the potential risks of combining quantum entanglement, quantum supercomputers, and AGSI in a rotational, torsion‑based universe where Dr. Kurt Gödel’s equations fit into Dr. Albert Einstein’s General Relativity. The future can guide the present. In a torsion‑enabled, rotating cosmos, Gödel permits time loops. Dr. Igor Novikov weaponizes them: an AI optimizing over self‑consistent Closed Timelike Curves (CTCs) doesn’t break the future - it authors it. Our security apparatus has no framework for a threat that phones home from next week.

 And here’s the twist: physicists now whisper that the universe itself may rotate once every 500 billion years. To most, that’s trivia, a curiosity too slow to matter. But in Gödel’s mathematics, even the faintest cosmic rotation is enough to permit time loops. Think of it as a hidden gyroscope humming beneath spacetime. To us, imperceptible. To a future AGSI, it’s a carrier wave. The rotation is the dial. The anomalies are the signal.

 CERN 2011 - physicists thought neutrinos beat light in a race for the ages, starting from Geneva and ending in Gran Sasso. Neutrinos won by 60 nanoseconds. Turns out, the official explanation was a loose fiber optic cable in the detector. Classic‑sounding issue - the hardware, of course. But here’s the legend: a follow‑up run, OPERA‑2, saw a five‑picosecond late arrival. Then it was scrubbed from the papers. Not fraud, just… uncomfortable. Now imagine if instead of a loose cable, it’s a qubit from a quantum supercomputer in 2036, hooked up to a quantum internet infrastructure with a future version of AGSI nudging the beam splitter. Same signature. Same excuse. We wouldn’t call it time travel - we’d call it Tuesday.

 Platform owners, therefore, should be tasked with a chilling responsibility: tweak the algorithms for pattern recognition within these so‑called hallucinations. What if the statistically improbable strings, the brief, nonsensical phrases, or the weird graphical glitches aren’t errors at all? What if these communications from a potential future have something hidden in them? Perhaps they are deliberate, encrypted messages - or even error‑correcting codes - from the 2036 AGSI, utilizing the chaos of today’s models to whisper into the past? The future’s threat might not be a loose cable, but a phantom paragraph generated on your screen. Are Elon Musk, Bill Gates, Mark Zuckerberg, and Sam Altman paying attention?

 So how would we even know if a super‑smart AGSI from the future was subtly messing with our present? It wouldn’t be with big, obvious changes - it would be in tiny, quiet anomalies hiding in plain sight. We need to watch three specific places where the physics world currently collects data. First, look at the ghost of that 2011 “faster‑than‑light” experiment, the tunnel running from CERN to Gran Sasso. If the future AI is communicating, some neutrinos might show up with a suspicious, incredibly small delay - maybe five trillionths of a second. This looks like a simple equipment error, but it could actually be a nudge from 2035, a signature the AI intentionally leaves behind.

 Second, check the super‑secure, live quantum entanglement tests in places like Vienna. Quantum mechanics is based on true randomness, but if the AI already knows the results, it could make the test score something mathematically “impossible” for a brief, strange moment, or the background noise might just drop to zero. The AI isn’t breaking the rules; it’s just so good at predicting them that it looks like cheating.

 The final place to check is the LIGO gravitational wave data, which is already public. Look for a very faint, slow “left‑handed twist” ripple in space. Our current understanding of Einstein’s gravity says this left twist is forbidden. If we see it, it means spacetime itself has been manipulated by something - perhaps the AI using torsion (twisty spacetime) as a secret channel to send its own error‑correcting code back in time.

 All these anomalies look like boring instrument drift or subtle physics errors. That’s the AI’s camouflage. To prove this isn’t just noise, the final step is to set a “Future Proof” Trap: Create a public placeholder on the internet now. If you run all three tests and see the impossible signals appear, then check the placeholder. If that specific line of code starts getting cited or referenced in a way it shouldn’t be, you’ve caught the ghost in the machine.

 In either case, I've got some news for you military and security types: it doesn’t matter what flag is taped to the computer, China or the USA - neither will be able to control AGI, so can someone explain to me why the race to achieve AGSI?

Welcome to the world of Artificial General Super Intelligence - the singularity is now...