Few companies and even fewer consumers will adopt OpenAI's o3 at wide scale in 2025 because of price and robustness concerns.

Judge this over 2025, for the model OpenAI released under the name o3. The claim is that few companies, and even fewer consumers, adopted o3 at wide scale, because of price and robustness concerns. TRUE: usage data, enterprise surveys, API share data or credible reporting show o3 was not widely adopted by companies or consumers in 2025, and price or cost relative to reliability is cited as a reason. MOSTLY TRUE: adoption was limited, but price or robustness was only one of several reasons, or a minor one. An example is o3 being quickly superseded by GPT-5. MIXED: adoption was limited, but the evidence attributes this to other causes, not price or robustness. Or: company adoption was limited while consumer use was wide, or the reverse. MOSTLY FALSE: o3 saw substantial adoption across many companies or a large consumer user base, with only some reported hesitancy about price or reliability. FALSE: o3 was widely adopted at scale by many companies and consumers in 2025, for example as a heavily used API model or a broadly used ChatGPT model, with no notable price or robustness barriers.

Set before the research began.

1. E1 Introducing GPT-5 OpenAI · 7 August 2025

   “GPT‑5 is the new default in ChatGPT, replacing GPT‑4o, OpenAI o3, OpenAI o4-mini, GPT‑4.1, and GPT‑4.5 for signed-in users.”

2. E2 OpenAI's new reasoning AI models hallucinate more | TechCrunch TechCrunch · 18 April 2025

   “OpenAI found that o3 hallucinated in response to 33% of questions on PersonQA, the company’s in-house benchmark for measuring the accuracy of a model’s knowledge about people.”

3. E3 Introducing GPT-5 OpenAI · 7 August 2025

   “we removed all the images from the prompts of the multimodal benchmark CharXiv, and found that OpenAI o3 still gave confident answers about non-existent images 86.7% of the time, compared to just 9% for GPT‑5.”

4. E4 O3 is 80% cheaper and introducing o3-pro OpenAI Developer Community · 10 June 2025

   “Thanks for the update on the o3 model pricing — the 80% reduction is a welcome change.”

5. E5 Introducing GPT-5 | OpenAI OpenAI · 7 August 2025

   “GPT‑5 is the new default in ChatGPT, replacing GPT‑4o, OpenAI o3, OpenAI o4-mini, GPT‑4.1, and GPT‑4.5 for signed-in users.”

6. E6 Introducing GPT-5 | OpenAI OpenAI · 7 August 2025

   “Across all of these benchmarks, “GPT‑5 thinking” shows a sharp drop in hallucinations—about six times fewer than o3—marking a clear leap forward in producing consistently accurate long-form content.”

7. E7 How 100 Enterprise CIOs Are Building and Buying Gen AI in 2025 · AI research paper excerpts and notes Inquiring Lines (summary of an Andreessen Horowitz (a16z) enterprise CIO survey) · June 2025

   “GPT-4o is the model most deployed to production, while OpenAI o3 has generated significant interest as reasoning models are more integrated into production use cases.”

8. E8 OpenAI’s o3 price plunge changes everything for vibe coders InfoWorld · 18 June 2025

   “On June 10, OpenAI slashed the list price of its flagship reasoning model, o3, by roughly 80% from $10 per million input tokens and $40 per million output tokens to $2 and $8, respectively.”

9. E9 OpenAI’s o3 price plunge changes everything for vibe coders InfoWorld · 18 June 2025

   “Until last week, that deliberation was priced like a Jeff Bezos-style turducken superyacht.”

10. E10 How 100 Enterprise CIOs Are Building and Buying Gen AI in 2025 · AI research paper excerpts and notes Inquiring Lines (summary of an Andreessen Horowitz (a16z) enterprise CIO survey) · June 2025

    “for most tasks, all the models perform well enough now—so pricing has become a much more important factor.”

No usage data, API share figures or adoption surveys in the evidence measure how widely companies or consumers used o3 in 2025. The a16z CIO survey summary says GPT-4o was the model most deployed to production, but also that o3 'generated significant interest' as reasoning models moved into production (E7). That does not show adoption was limited. There is evidence of robustness problems, such as hallucination rates (E2, E3, E6). OpenAI cut o3's list price by about 80% in June (E4, E8), and GPT-5 replaced o3 as the ChatGPT default in August (E1). None of these items shows how widely o3 was adopted, or ties any limit on its use to price or robustness, so the claim cannot be settled.