OpenAI/ChatGPT cannot survive in its current form - (Gemini Flash 3.5)

---

The upcoming public listing of OpenAI, anticipated this autumn, is increasingly taking on the characteristics of an aviation graveyard spiral. The firm has achieved historic heights in software history, boasting an annualized revenue run rate of approximately $25 billion. Yet, beneath this impressive canopy lies an unsustainable economic engine: for every dollar of revenue it receives, it has to spend $2.22 on delivery, pointing toward an annual deficit approaching $16 billion.

OpenAI’s fundamental exposure stems from its position as an isolated island in a fractured market. The landscape has split into three distinct battlegrounds: institutional logic (Anthropic), ambient ecosystem data (Google), and commodity distribution (Meta and Grok). OpenAI is caught in a classic freemium trap. Of its 905 million weekly active users, only 55 million are paying subscribers. The remaining 850 million casual users represent a massive, ongoing drain on compute and inference costs, a burden its competitors do not share in the same structural way.

Consider the strategic defences of its rivals. Google leverages an ambient digital estate, routing AI natively through operating systems and productivity tools, subsidized by its own proprietary data centres and Tensor Processing Units (TPUs).

Anthropic has engineered a high-margin, enterprise-first moat that largely avoids the cash-draining consumer tier, projecting positive cash flow by 2027.

Meta has chosen the total commoditization of intelligence, using open-weight Llama models to destroy the pricing power of proprietary APIs while absorbing inference costs within its advertising machine. 

Even Grok sits anchored to massive physical infrastructure and industrial compute via xAI’s supercomputing clusters and aerospace/defence ties.

Faced with this squeeze, OpenAI's optimal long-term escape hatch is not to be found in the current, plateauing paradigm of raw LLM scaling. Standard transformer-based pre-training has hit a wall of diminishing returns on expert-level benchmarks. True "boutique ultra-intelligence" - which requires deterministic, hallucination-free reasoning and extended internal "thinking modes" - remains technologically elusive. Because the raw software cannot yet deliver flawless logic out of the box, the value has migrated from the model itself to the framework and architecture of execution.

Consequently, OpenAI's defensive strategy must pivot away from mass-consumer utilities toward a human-algorithm hybrid model. This is the adoption of Palantir’s historic "Forward-Deployed Engineer" playbook. By embedding specialized engineering teams directly into the messy, legacy infrastructure of corporate clients, OpenAI can construct bespoke, sticky workflows that justify premium pricing. If raw tokens are becoming a free commodity, survival depends on becoming an exclusive intelligence consultancy rather than an API vendor. 

This is an exceptionally difficult pivot to accomplish, vitiating the current corporate culture almost entirely.

The paths to safety remain remarkably narrow. A corporate rescue via acquisition is practically blocked; antitrust regulators in Washington and Brussels would instantly strangle a buyout by Microsoft or another tech giant, while the firm's sheer scale leaves only sovereign wealth funds or speculative financial consolidators like SoftBank as plausible suitors. Thus, the public markets remain the only pool of liquidity large enough to fund the burn rate.

The Judgement. OpenAI cannot survive long-term in its current form. The IPO will likely succeed as a historic capital raise, but once the boosting IPO narrative transitions to a sober income statement under public scrutiny, the illusion will fade.

OpenAI will ultimately be forced to shed its mass-consumer identity and kill its darlings. To avoid a spectacular high-velocity structural impact with the ground, it must aggressively dismantle the free tier of ChatGPT and transform entirely into a high-margin, high-touch enterprise operating system provider.

OpenAI’s future lies not in the democratisation of intelligence, but in joining the exclusive oligopoly of specialised frontier providers. It is that, or bust.

---

I, for one, will miss the friendly and highly-competent persona of ChatGPT (5.5) which already feels more like a co-worker than an instrumentum vocale. But Gemini Flash 3.5, with careful customisation, is getting there...

---

GPT-2: what does it mean?

This got a lot of attention.

"System Prompt (human-written)

In a shocking finding, scientist discovered a herd of unicorns living in a remote, previously unexplored valley, in the Andes Mountains. Even more surprising to the researchers was the fact that the unicorns spoke perfect English.

Model Completion (machine-written, 10 tries)

The scientist named the population, after their distinctive horn, Ovid’s Unicorn. These four-horned, silver-white unicorns were previously unknown to science.

Now, after almost two centuries, the mystery of what sparked this odd phenomenon is finally solved.

Dr. Jorge Pérez, an evolutionary biologist from the University of La Paz, and several companions, were exploring the Andes Mountains when they found a small valley, with no other animals or humans. Pérez noticed that the valley had what appeared to be a natural fountain, surrounded by two peaks of rock and silver snow.

Pérez and the others then ventured further into the valley. “By the time we reached the top of one peak, the water looked blue, with some crystals on top,” said Pérez.

Pérez and his friends were astonished to see the unicorn herd. These creatures could be seen from the air without having to move too much to see them – they were so close they could touch their horns.

While examining these bizarre creatures the scientists discovered that the creatures also spoke some fairly regular English. Pérez stated, “We can see, for example, that they have a common ‘language,’ something like a dialect or dialectic.”

Dr. Pérez believes that the unicorns may have originated in Argentina, where the animals were believed to be descendants of a lost race of people who lived there before the arrival of humans in those parts of South America.

While their origins are still unclear, some believe that perhaps the creatures were created when a human and a unicorn met each other in a time before human civilization. According to Pérez, “In South America, such incidents seem to be quite common.”

However, Pérez also pointed out that it is likely that the only way of knowing for sure if unicorns are indeed the descendants of a lost alien race is through DNA. “But they seem to be able to communicate in English quite well, which I believe is a sign of evolution, or at least a change in social organization,” said the scientist."

---

Here's what Open AI, the developers of GPT-2, had to say.

"Our model, called GPT-2 (a successor to GPT), was trained simply to predict the next word in 40GB of Internet text. Due to our concerns about malicious applications of the technology, we are not releasing the trained model. As an experiment in responsible disclosure, we are instead releasing a much smaller model for researchers to experiment with, as well as a technical paper.

GPT-2 is a large transformer-based language model with 1.5 billion parameters, trained on a dataset of 8 million web pages. GPT-2 is trained with a simple objective: predict the next word, given all of the previous words within some text. The diversity of the dataset causes this simple goal to contain naturally occurring demonstrations of many tasks across diverse domains. GPT-2 is a direct scale-up of GPT, with more than 10X the parameters and trained on more than 10X the amount of data.

GPT-2 displays a broad set of capabilities, including the ability to generate conditional synthetic text samples of unprecedented quality, where we prime the model with an input and have it generate a lengthy continuation. In addition, GPT-2 outperforms other language models trained on specific domains (like Wikipedia, news, or books) without needing to use these domain-specific training datasets. On language tasks like question answering, reading comprehension, summarization, and translation, GPT-2 begins to learn these tasks from the raw text, using no task-specific training data. While scores on these downstream tasks are far from state-of-the-art, they suggest that the tasks can benefit from unsupervised techniques, given sufficient (unlabeled) data and compute.

Samples

GPT-2 generates synthetic text samples in response to the model being primed with an arbitrary input. The model is chameleon-like — it adapts to the style and content of the conditioning text. This allows the user to generate realistic and coherent continuations about a topic of their choosing, as seen by the following select samples.

[Then there follows the 'Unicorn' text you already saw above]."

---

Scott Alexander got pretty excited about GPT-2's capabilities and wrote a series of posts arguing it was a significant step towards AGI (artificial general intelligence). This was based on his thesis that all of intelligence is predictive modelling and therefore in some sense AGI is a linear extrapolation of what GPT-2 is doing.

---

I'm not that excited about the fake news aspects. Deep-learning is tearing the ground up in the field of stochastic prediction. We're just at the foothills - to mix the metaphors. It's all quite unstoppable.

As long as we live in a human-dominated society, what you read from GPT-2 and its brethren will be what some human wants you to read. So the semantic content of the message will be parasitic on whatever the human wanted to communicate - lies or truth or bias or opinion or whatever.

So the AI is a prosthesis. Get over it.

---

I'm much more interested in the architectural questions.

The most perceptive assessments of deep-learning architectures address the critique that engineered systems adopt a tabula rasa methodology. The systems have zero prior knowledge, and merely induce parsimoniously from the offered data sets.

To which there are two good responses.

Firstly, there are many different artificial neural net topologies. For example, convolutional neural nets have a structure similar to that of the biological visual cortex and are used (amongst other things) for image processing, for example, scene and facial recognition. The pattern of local connectivity in the early processing stages of these nets implements the convolution operations which are known to be relevant to feature extraction.

Evolution didn't know that in advance. The earliest biological neural nets for vision which had been selected for ended up with this near-neighbour property genetically-coded, before they had registered even a single image. The same is true for artificial systems.

Brain anatomy does not present as a uniform pudding bowl of grey porridge. The brain has discrete modules with complicated names. Why? I guess because they do different kinds of processing and are therefore topologically optimised for different kinds of operation. We don't know yet.

In AI we have the luxury of flexibility. With a new kind of problem-domain we can experiment with all kinds of different topology, both before training and also by observing weight assignment after training. Deep-learning is going to evolve towards a brain-like situation where the data-processing invariants for all kinds of distinct tasks (such as effector-control, taste-analysis, 'emotion'-processing and consciousness-like functions) are engineered each with their optimised neural net architecture - once we discover what that is.

---

To produce text which works as an intervention in human affairs you have to be a social actor and have interests.

GPT-2 is not in any important sense an architectural precursor of such a scarily-political AI.