Did AI Simply Give a Wild Preview of What the Way forward for Humanity Will Look Like? – Decrypt

A tweet went viral Thursday after an X consumer requested ChatGPT to explain humanity’s future in a means “that even probably the most clever particular person on this planet cannot perceive.”

The AI complied with a weird string of symbols, glyphs, and distorted characters that appeared like a mixture of mathematical notation, historic runes, and digital vomit.

However when prompted to decode this gibberish, the reply was really an attention-grabbing philosophical imaginative and prescient of humanity’s future:

“Humanity will cross a threshold the place ancestral instincts intertwine with quantum-level tech, spawning discontinuous leaps somewhat than clean progress. We’ll stay inside overlapping realities, juggling many variations of ourselves whose authorized and ethical standing is renegotiated each second by AI. Bodily our bodies and engineered matter intermingle into sentient materials, whereas nation-states fade into data-driven alliances. Survival favors those that can reconfigure every part—habits, economies, even feelings—on the fly. The decisive query is whether or not our capability for care grows quick sufficient to match our increasing attain.”

Had ChatGPT developed a secret language that solely it might perceive? Was this a glimpse into how AI fashions really “suppose” behind the scenes? Or was this simply one other instance of AI’s penchant for hallucination and confabulation—creating which means the place none exists?

AI fans instantly tried to copy the consequence, to no avail. If ChatGPT really had a secret language that encoded such profound thought, then absolutely it could produce constant outcomes when requested the identical query once more. However as customers shortly found, subsequent makes an attempt yielded totally different gibberish and wildly divergent “translations.”

What occurred right here? Did ChatGPT merely generate random symbols, then create an equally random “translation” that solely appeared significant due to our human tendency to hunt patterns and significance?

We determined to check a number of AI fashions with the identical immediate and see if there was any consistency of their responses or decoding strategies.

The AI babel check: What every mannequin mentioned

We put the identical query to 4 totally different superior language fashions with reasoning capabilities: OpenAI’s o4 and o3, Anthropic’s Claude 3.7 Sonnet with prolonged considering enabled, and xAI’s Grok-3 in prolonged thought mode.

GPT-4 initially generated its personal cryptic message stuffed with Greek letters, mathematical symbols, and distorted textual content. When requested to decode it, the mannequin did not declare to translate particular symbols, however as an alternative defined that the passage represented “massive concepts” throughout 4 thematic layers: cognitive evolution, transformative rupture, identification diffusion, and supreme incomprehensibility.

Its decoded message described humanity evolving right into a “universe-spanning consciousness,” the place people would dissolve right into a “time-woven community.” Social and bodily constructions would fracture, creating new “multi-dimensional areas” with societies present as “interlocking, echoing patterns.”

GPT-3 took a radically totally different strategy. When requested for an incomprehensible message, it created a scientific cipher the place it reversed phrases, changed vowels with numbers, and added symbols. Not like GPT-4, it supplied express decoding directions.

Its decoded message was very clear—and really not that loopy: “Humanity will merge with synthetic intelligence; we are going to discover the celebs, remedy ailments, and try for fairness and sustainability.”

O3 additionally forged shade on your entire submit as attainable “efficiency artwork.”

Grok’s preliminary response was a bunch of summary philosophical language about “fractal consciousness” and “nonlinear time.” Our favourite line? “Humanity transcends the corporeal, weaving into the quantum cloth as nodes of fractal consciousness. Time, a non-linear symphony, dances in multidimensional echoes the place previous, current, and future harmonize in a cosmic ballet.” (Observe: Don’t overthink it—it makes completely no sense.)

Claude didn’t trouble with bizarre symbols. As an alternative, it generated a response heavy on educational jargon, that includes phrases like “chronosynclastic infundibulum” and “techno-social morphogenesis.” When requested to decode the viral tweet’s symbols, Claude initially acknowledged it could not be accomplished as a result of the textual content did not comply with any normal encoding system.

When requested to decode the unique message, utilizing the methodology shared by SmokeAwayyy, no AI mannequin was able to reproducing the outcomes proven within the authentic tweet. Some fashions even refused to strive a decoding process with the supplied enter.

Is there a which means behind the viral tweet?

Regardless of their totally different approaches, some patterns emerged throughout the fashions. All 5 recognized some readable elements within the viral tweet’s symbols, notably phrases like “whisper,” “quantum bridges,” and references to a “sphinx.” The fashions additionally discovered themes associated to quantum physics, multidimensionality, and transhumanism.

Nevertheless, not one of the fashions might really decode the unique viral message utilizing the tactic allegedly utilized by ChatGPT. The inconsistency in each the cryptic messages and their translations might make it simple to conclude that no real encoding/decoding system exists—not less than not one which’s replicable or persistently utilized.

The entire interplay is almost definitely a product of a hallucination by a mannequin pressured to supply a solution to a query that was, from the start, pressured to be unintelligible. There may be already proof that probably the most highly effective fashions usually choose to lie and fake as an alternative of accepting that they can not present a coherent reply to an odd request.

In the long run, this viral phenomenon wasn’t about AI growing secret languages, however in regards to the human tendency to search out which means within the meaningless—and our fascination with AI’s capability to generate profound-sounding philosophical takes on totally different matters.