ChatGPT Brand Analysis: Still the Category Name, but the Lead Is Narrowing

ChatGPT is still the category-defining AI brand, but its lead in the United States is narrowing, and its main brand risk is trust rather than awareness. The best public evidence is a tracking survey of US adults who used generative AI in the past 30 days, covering February to July 2026.

The scorecard

Measure (US, July 2026)	ChatGPT	Gemini	Copilot	Claude
Preferred assistant	33.7%	18.2%	8.2%	6.1%
Accuracy, net score	30.3	25.3	13.9	14.4
Personal impact, net score	35.3	25.9	18.4	14.0
Preference change, Feb to Jul 2026	-5.9 pts	+6.8 pts	+3.0 pts	+0.8 pts

Net scores are positive minus negative perceptions.

ChatGPT leads every level measure. The last row is the warning: it lost 5.9 points of preference in five months while Gemini gained 6.8.

Positioning: everyone's assistant

ChatGPT is positioned as the general-purpose assistant for everyone, from a free student account to an enterprise seat. That breadth is the brand's strength and its constraint. It is the default choice but not the specialist choice in coding, search or enterprise security, where Claude, Perplexity and Microsoft Copilot each claim a niche. The survey finds ChatGPT scores 10.9 points higher among knowledge workers than among other workers, and Claude 7.1 points higher, so the two brands compete for the same professional audience.

Brand architecture: three layers, plus plan names

Three layers coexist: the corporate brand (OpenAI), the product brand (ChatGPT), and model names. The current family is GPT-5.6, with the variants Sol, Terra and Luna; older GPT-5.4 and 5.5 models were retired on 31 August 2026. Plan names add another layer: Free, Go, Plus, Pro at $100 and $200, Business and Enterprise. That is a lot for a mainstream user to hold in their head, and it is the likely source of confusion about which model and which limit applies to whom.

The brand in India

OpenAI has localised the brand through price (Go at Rs 399, then free for a year), payments (UPI), language (the IndQA benchmark across 12 Indian languages) and institutions (Edu licences at IIM Ahmedabad and AIIMS). It has a New Delhi office, with Mumbai and Bengaluru planned. Sam Altman's line that India has the largest number of students using ChatGPT anywhere makes the brand's Indian identity a learning-and-work tool for young users.

Three brand risks

Trust. The abrupt GPT-4o retirement in January 2026 and the loss of senior safety leaders fed a sustained consumer backlash.

Perceived throttling. As caps tighten and heavy users are moved to lighter models, paying users may read it as a broken promise even where the cause is capacity.

Erosion at the margin. Gemini's distribution through Android, Search and telecom bundles is designed to erode a lead like this one.

What this evidence cannot tell us

The survey data covers the US and UK. I found no equivalent brand-tracking study for India, so the Indian brand view rests on OpenAI's actions and adoption data, not on measured perception. Until someone measures it, the most reliable statement is that ChatGPT's Indian brand is built on reach and price, and its US brand is being tested on trust.