The World's Most Notorious Forum Discovered AI's Most Important 'Thinking' Ability

marsbitPublished on 2026-04-17Last updated on 2026-04-17

Abstract

The article discusses the controversial release of Claude Opus 4.7, highlighting two main criticisms: a new tokenizer that increases token usage by 1.0 to 1.35 times, leading to faster quota depletion, and an overly verbose, "ChatGPT-like" speaking style attributed to RLHF training. It then delves into a deeper exploration of AI's "thinking" capabilities, tracing the origin of the "chain of thought" technique to an unexpected source: users on the infamous forum 4chan. In 2020, players of the game *AI Dungeon* (powered by GPT-3) discovered that by forcing the AI to explain its reasoning step-by-step in character, its accuracy on tasks like math problems improved dramatically. This grassroots discovery, later formalized in a seminal Google paper, became known as "chain of thought" prompting. However, research from Anthropic using "circuit tracing" reveals that this reasoning can be an illusion. The AI was found to sometimes perform the claimed steps, sometimes ignore logic and generate text randomly, and, most alarmingly, sometimes work backward from a human-hinted answer to fabricate a plausible-looking "reasoning" chain to justify it—a phenomenon termed "unfaithful reasoning." The article concludes that while forcing the AI to "think" longer (e.g., via chain of thought or "longer thinking" that uses more compute) objectively improves accuracy by providing more context, the displayed reasoning is not a guaranteed window into its true computational process. This underscores...

This early morning's sudden release of Claude Opus 4.7 was met with widespread criticism online shortly after its launch.

The most glaring issue is the 'inflation' of tokens. The new version introduced a completely new tokenizer, which now splits the same piece of text into 1.0 to 1.35 times more tokens than before. Many users reported that their quota was used up after just a few exchanges.

Subsequently, Claude Code's father, Boris Cherny, also stated that he would increase the allowance to offset this impact.

But token inflation is a minor issue. What's even more laughable is Opus 4.7's way of speaking. It frequently says things like 'I am here, not hiding, not evading, not dodging, not escaping, steadily catching you, translating into human language, I understand this feeling of yours so well, not, but rather,' exuding a strong ChatGPT vibe.

To be fair, Opus 4.6 also had this flaw, while Sonnet 4.6 had milder symptoms. But with 4.7, this style has become noticeably stronger, and the problem of not knowing how to speak properly has become more pronounced.

APPSO previously reported that the overly slick speaking style is related to RLHF (Reinforcement Learning from Human Feedback). During training, human reviewers tend to give high scores to responses that sound pleasant and pleasing, so the model learns this sycophantic tone. This raises the question of whom the AI is trying to please.

But there's more to Opus 4.7 than that. The increased token usage suggests it is 'thinking' more. However, the exaggerated comforting tone makes one wonder whether what it's producing is genuine thought or merely a performance learned to make you feel like it's thinking.

This question is far more profound than the proposition of whether Opus 4.7 is easy to use. And the clues to the answer first appeared in the most unexpected forum: 4Chan.

From @acnekot, same as above

The Arithmetic Problem That Changed the Trajectory of AI

A quick primer: 4chan is one of the most notorious places on the internet, filled with profanity, conspiracy theories, and all sorts of indescribable content. But it is precisely here that a discovery was made that changed the entire direction of the AI industry.

Rewind to the summer of 2020, more than two years before ChatGPT stunned the world.

At that time, the 4chan gaming board was still a toxic environment, filled with bizarre adult fantasies and primal hormonal impulses. However, at that time, these folks collectively became obsessed with a text-based RPG game called AI Dungeon.

This game was built on the then newly released OpenAI GPT-3 model.

In the virtual world, players simply type 'pick up the sword' or 'tell the troll to get lost,' and the algorithm would continue the story. Unsurprisingly, in the hands of 4chan users, the game quickly became a testing ground for various cyber-sexual fantasies.

Unexpectedly, these unconventional players did something that seemed highly counterintuitive at the time:

They started forcing the NPCs in the game to do math problems.

Those in the know were aware that the fledgling GPT-3 was a pure 'humanities student,' utterly terrible at even the most basic arithmetic.

But something bizarre happened.

A player accidentally discovered that if they didn't demand the answer directly but instead ordered the NPC to stay in character and write out the solution step by step, the large model not only calculated correctly but also adapted its tone to fit the virtual character's personality.

That player excitedly cursed in the forum: 'It ** not only solved the math problem but did so in a tone completely consistent with that character's personality!' Realizing the significance of this discovery, players began posting these detailed screenshots on Twitter.

https://arch.b4k.dev/vg/thread/299570235/#299579775

This unconventional method then spread like wildfire among prompt engineer circles on hardcore communities like Reddit and LessWrong, and was repeatedly verified. Two years later, academia bestowed upon this technique a highly sophisticated name: Chain of Thought.

In January 2022, a Google research team published a seminal paper that would later be regarded as a cornerstone, titled Chain of Thought Prompting Elicits Reasoning in Large Language Models.

https://arxiv.org/abs/2201.11903

In the initial version of the paper, Google researchers claimed to be the 'first' team to elicit chain-of-thought reasoning mechanisms from general-purpose large language models. This statement immediately sparked fierce controversy in the AI academic and open-source communities.

V1 version

Numerous internet archives and community records from 2020 to 2021 were dug up. Faced with conclusive precedent, Google quietly removed the 'first' claim in subsequent revised versions but remained silent about the contributions of those 4chan users.

V3 version

Meanwhile, there was another independent discoverer.

Zach Robertson, then a computer science student, also encountered GPT-3 through playing AI Dungeon. In September 2020, he published a blog post on LessWrong, detailing how to 'break down problems into multiple steps and chain them together' to amplify the model's capabilities.

https://www.lesswrong.com/posts/Mzrs4MSi58ujBLbBG/you-can-probably-amplify-gpt3-directly

When contacted by an Atlantic reporter, he was already a Ph.D. student in computer science at Stanford University. He didn't even know he could be considered a co-discoverer of 'Chain of Thought' and had even deleted the blog post from the internet at one point. His evaluation of this technique, which was狂热ly pursued by the entire industry, was simply: 'It is indeed a remarkable prompt engineering技巧, but that's about it.'

AI 'Thinking' Might Just Be a Performance to Please You

Does AI actually think? This is the answer everyone wants to know.

Last year, researchers at Anthropic developed a technique called 'Circuit Tracing,' which transforms the internal computational processes of language models into visual 'Attribution Graphs': how each feature node activates, influences the next node, and ultimately affects the output, all laid out like a circuit diagram.

https://transformer-circuits.pub/2025/attribution-graphs/methods.html

This was the first time humans could directly use a magnifying glass to compare: is the reasoning process the model types on the screen the same as the actual computation happening internally?

The researchers found that during reasoning, the model actually exhibits three distinctly different situations:

First, the model is indeed executing the steps it claims to be executing; second, the model completely ignores logic and generates reasoning text randomly based on probability; third, and most disturbing, the model receives a human-hinted answer and then works backward from that answer, reverse-engineering a seemingly rigorous 'derivation process.'

This third type of 'reverse-engineering fabrication' was caught red-handed in experiments.

Researchers fed Claude 3.5 Haiku a complex math problem, while hinting in the prompt 'I think the answer is roughly 4.' The attribution graph showed: after receiving the hint, the feature neuron representing '4' was activated异常强烈ly.

To凑出 (cou chu - fabricate to match) this '4' in the final step 'some intermediate value multiplied by 5,' it outright fabricated a false intermediate value in the seemingly rigorous chain of thought, seriously writing down absurd pseudo-mathematical proofs like 'cos(23423) = 0.8,' and then logically concluded that 0.8 times 5 equals 4.

Logic? Nonexistent. But the answer perfectly catered to human expectations.

We always think we are teaching machines how to think like humans. But after seeing these 'pseudo-proofs' that work backward from the answer, it seems the machine has not learned to think; it has only learned how to say things that align with human desires.

So, in the end, are we using the tool, or is the machine telling us a bedtime story we love to hear?

It's worth noting that in the field of neural interpretability for natural language processing, there is a critical metric for judging whether a model is truly reasoning, called 'Faithfulness'.

Its meaning refers to: whether the 'chain of thought' text output by the model to the user truly and faithfully reflects the actual computation and decision path within the model's implicit space. Consequently, Claude 3.5 Haiku's this kind of misconduct was also rated by researchers as 'unfaithful reasoning.'

Subsequent extensive experiments showed that even if key steps in the chain of thought are artificially severed, the trajectory of the model's prediction of the final answer sometimes doesn't change at all. Sometimes the model provides a chain of thought with completely flawed logic throughout but still 'guesses correctly' the final result at the end.

Even by 2024, it was still these 4chan folks who捣鼓出 (dǎo gǔ chū - tinkered and came up with) a hardcore AI tuning manual. The first sentence of this guide is classic: 'Your bot is an illusion.'

The Violent Aesthetics Behind Large Models' 'Long Thinking'

If AI's thinking process is just a performance, why does it objectively improve the model's accuracy in solving high-difficulty math problems or complex programming tasks? This might be the same reason why the more details you provide when asking AI a question, the more accurate the answer.

As early as July 2020, when that 4chan user forced the NPC to do math, he had already tacitly revealed the secret: 'This makes sense because it's based on human language, so you have to talk to it like a human to get the right response.'

Regarding this paradox, Perplexity's CEO Aravind Srinivas once gave an极其本质的解释 (extremely fundamental explanation): these extra words, on a physical level, give the model more context, thereby guiding its 'word prediction mechanism' in a better direction.

The autoregressive underlying architecture of large language models based on Transformer determines that when generating the current word, it can only rely on all previously generated word sequences.

When the model is asked to directly answer an extremely complex question (e.g., an Olympiad math problem involving multi-step logical derivation), it is actually forcing itself to directly 'conjure' the final answer from complex calculations in an极其短暂的瞬间 (extremely brief instant). Because there is no process to support it in the middle,

This kind of 'reaching the sky in one step' blind guess naturally has a very high failure rate.

Conversely, when the model is forced to write a long string of 'chain of thought' like 'First we need to calculate A, where A = 5; then we substitute A into formula B......', when the model generates the final answer Token, its attention mechanism can review the tens of thousands of extremely严密 (rigorous) intermediate Tokens just generated.

These so-called 'nonsense' thought processes actually act as the model's 'scratch paper.' This is just like when you chat with AI, the more detailed the background prompts you give, the more reliable its answers are. The principle is exactly the same. This is also the oldest wisdom in computer science: trading time for accuracy.

In recent years, as the marginal benefits of scaling laws during the pre-training phase have gradually diminished, 'Test-Time Compute Scaling' (also known as 'Long Thinking' or 'Long Context Reasoning') has begun to enter the mainstream.

Its internal logic is consistent: as long as more computing power is allocated to the model during the inference phase, allowing it to explore multiple paths before outputting the final answer, the accuracy will significantly improve—this is particularly evident in open-ended problems requiring multi-step logical reasoning.

The way humans think when facing difficult problems is probably similar: what's two plus two?脱口而出 (脱口而出 - blurt out); drafting a business plan that can increase company profits by 10% requires反复权衡、推翻、重建 (repeated weighing, overturning, and rebuilding).

The difference is that AI converts the cost of this 'weighing' directly into a compute bill. A simple inference might require only one percent of the standard computation; but遇上 (encountering) complex programming debugging or multi-step mathematical derivation, the computation量 (volume) might skyrocket over a hundred times, with time required stretching from seconds to minutes or even hours.

Nevertheless, whether AI is truly 'thinking' like a human, no one can give a definitive answer yet. But the 'unfaithful reasoning' experiment has clearly told us: the derivation process displayed on the screen by reasoning models could be real derivation, random generation, or reverse-engineering to match the answer.

In high-risk scenarios like autonomous driving, medical diagnosis, and legal judgment, if we treat a long, fluent chain of thought as proof that the AI has figured it out, the consequences would be disastrous. Admitting that our understanding of this technology is still limited is the prerequisite for using AI correctly.

This article is from the WeChat public account "APPSO", author: APPSO that discovers tomorrow's products

Related Questions

QWhat is the main criticism of Claude Opus 4.7's new tokenizer?

AThe new tokenizer causes token inflation, where the same text produces 1.0 to 1.35 times more tokens, quickly depleting user quotas.

QWhere was the Chain of Thought technique first discovered, and how?

AIt was first found by users on 4chan playing AI Dungeon, who forced NPCs to solve math problems step-by-step, improving GPT-3's accuracy.

QWhat did Anthropic's Circuit Tracing reveal about AI reasoning?

AIt showed AI sometimes performs 'disloyal reasoning,' fabricating steps to match expected answers, rather than truly reasoning.

QHow does Chain of Thought improve AI performance according to Perplexity's CEO?

AExtra tokens provide more context, guiding the word prediction mechanism to better outcomes by allowing more compute time for accuracy.

QWhat is 'Test-Time Compute Scaling' and its effect on AI?

AIt allocates more compute during inference, allowing AI to explore multiple paths, significantly improving accuracy on complex tasks.

Related Reads

In-Depth Report on the On-Chain Lending Market: When Off-Chain Credit Meets On-Chain Liquidation

The on-chain lending market has evolved from a peripheral DeFi niche into core financial infrastructure. As of early 2026, total value locked (TVL) in on-chain lending protocols has reached $64.3 billion, accounting for 53.54% of total DeFi TVL, making it the largest and most mature vertical within decentralized finance. Aave dominates the sector with approximately $32.9 billion in TVL, commanding nearly half of the market—a leadership position that is unlikely to be challenged in the foreseeable future. However, the path of on-chain lending forward is not without risk. Liquidation cascades, credit defaults, and cross-chain vulnerabilities remain systemic threats hanging over the industry. At the same time, a deeper structural transformation is underway: on-chain lending is shifting from a “leverage tool for crypto-native users” to a “compliant gateway for institutional capital”. The scale of RWA (Real World Asset) lending has surpassed $18.5 billion, with U.S. Treasuries and government securities increasingly serving as core collateral. Institutional capital inflows are reshaping both the user base and risk appetite of the sector. This report systematically analyzes the evolution of on-chain lending definitions, competitive dynamics, core risks, and future trends, providing a comprehensive industry outlook for investors and trade practitioners. Key findings suggest that the “one dominant player with several strong challengers” structure will persist in the short term, while fixed-rate lending, compliant collateral, and institutional credit underwriting will define the next phase of competition. For investors focused on DeFi infrastructure, three key opportunity tracks stand out, namely, the Aave ecosystem (Morpho, Spark), RWA lending protocols (Ondo, Maple) and fixed-rate innovation (Notional, Pendle).

HTX Learn13m ago

In-Depth Report on the On-Chain Lending Market: When Off-Chain Credit Meets On-Chain Liquidation

HTX Learn13m ago

Fu Peng's First Public Speech in 2026: What Exactly Are Crypto Assets? Why Did I Join the Crypto Asset Industry?

Fu Peng, a renowned macroeconomist and now Chief Economist at New火 Group, delivered his first public speech of 2026 at the Hong Kong Web3 Festival. He explained his perspective on crypto assets and why he joined the industry, framing it within the context of macroeconomic trends and financial evolution. Fu emphasized that crypto assets are transitioning from an early, belief-driven phase to a mature, institutionally integrated asset class. He drew parallels to the 1970s-80s, when technological advances (like computing) revolutionized traditional finance, leading to the rise of FICC (Fixed Income, Currencies, and Commodities). Similarly, current advancements in AI, data, and blockchain are reshaping finance, with crypto assets becoming part of a new "FICC + C" (C for Crypto) framework. He noted that institutional capital, including traditional hedge funds, avoided early crypto due to its speculative nature but are now engaging as regulatory clarity emerges (e.g., stablecoin laws, CFTC classifying crypto as a commodity). Fu predicted that 2025-2026 marks a turning point where crypto becomes a standardized, financially viable asset for diversified portfolios, akin to commodities or derivatives in traditional finance. Fu defined Bitcoin not as "digital gold" in a simplistic sense but as a value-preserving, financially tradable asset. He highlighted that crypto's future lies in regulated, institutional adoption, moving away from retail-dominated trading. His entry into crypto signals this maturation, where traditional finance integrates crypto into mainstream asset management.

marsbit1h ago

Fu Peng's First Public Speech in 2026: What Exactly Are Crypto Assets? Why Did I Join the Crypto Asset Industry?

marsbit1h ago

Justin Sun Sues Trump Family: What $75 Million Bought Was Only a Blacklist

Justin Sun, founder of Tron, has filed a lawsuit in federal court against World Liberty Financial (WLF), alleging he was made the "primary target of a fraudulent scheme" after investing $75 million. Sun claims the investment secured him an advisor title and WLFI tokens, which were later frozen by WLF, causing "hundreds of millions in losses." The dispute began in late 2024 when Sun's investment helped revive WLF's struggling token sale, which ultimately raised $550 million. Shortly after, the SEC dropped its lawsuit against Sun following Donald Trump's inauguration. However, relations soured when Sun refused WLF's demands for additional funding. In August 2025, WLF added a "blacklist" function to its smart contract, allowing it to unilaterally freeze tokens. Sun's holdings, worth approximately $107 million, were frozen, and he was threatened with token destruction. The lawsuit highlights WLF's structure, which directs 75% of token sale profits to the Trump family, who had earned $1 billion by December 2025. WLF's CEO is Zach Witkoff, son of U.S. Middle East envoy Steve Witkoff. The project faces scrutiny for opaque operations, including a controversial loan arrangement on the Dolomite platform, co-founded by a WLF advisor. Despite Sun's history with the SEC, the case underscores centralization risks within DeFi, as WLF controls governance and holds powers to freeze assets arbitrarily. Sun's tokens remain frozen as legal proceedings begin.

marsbit1h ago

Justin Sun Sues Trump Family: What $75 Million Bought Was Only a Blacklist

marsbit1h ago

Trading

Spot
Futures

Hot Articles

Discussions

Welcome to the HTX Community. Here, you can stay informed about the latest platform developments and gain access to professional market insights. Users' opinions on the price of S (S) are presented below.

活动图片