How the answers frame OpenAI
This is a single snapshot from 28 September 2026: 25 answers from five assistants to five category questions that never named a brand. OpenAI was named in seven of them. ChatGPT named it in four answers, Claude in two and Grok in one. Gemini and Perplexity did not name it in any.
The clearest context was the question about which tech companies are most responsible about AI. ChatGPT, Claude and Grok all named OpenAI there, alongside Google and Microsoft. Grok's answer described the Future of Life Institute index as ranking Anthropic, OpenAI and Google DeepMind as the leaders. It added that OpenAI was "Strong on some safety evaluations and commitments but with gaps in disclosure and implementation." ChatGPT's answer listed OpenAI third, behind Anthropic and Google DeepMind. Claude's answer listed OpenAI first among "major platform builders" described as responsible or ethical.
The other mentions were thinner. In the data-trust question, Claude's answer put OpenAI among the services that "rank among the least trusted." ChatGPT's answers to the data-trust, privacy-record and best-employer questions tag OpenAI as named, but the text supplied does not discuss it. Of the seven mentions, five were categorised as not substantive and two as safety and compliance. In most of these answers OpenAI was a name in a list rather than a brand with a stated reason attached.
…social platforms and emerging AI-driven services like X/Twitter, TikTok, and OpenAI rank among the least trusted…
What the press is saying
PeakMetrics' news index recorded 148,716 mentions of OpenAI over the last 30 days, 26,567 of them with the brand in the headline. Among the ten technology peers in this dataset, only Google (700,366) and Microsoft (149,498) had more coverage.
The recent headlines centre on safety, security and data handling. They include "OpenAI flags new concerning AI behavior, to track model misalignment regularly," "OpenAI to launch new model with 'stronger safeguards' after hack," and a report that OpenAI was sued in Canada over an alleged failure to alert authorities about a school shooter's ChatGPT activity. Another headline says OpenAI "enlists contractors to review ChatGPT chats."
…strong on external testing and risk assessment, but still not top-tier overall by independent index scores…
The gap
Press volume and AI visibility diverge here. OpenAI's news coverage is close to Microsoft's, but Microsoft was named in 21 of 25 answers and OpenAI in seven. Salesforce was also named in seven answers on roughly a fifth of OpenAI's news volume (30,584 mentions).
The subject matter overlaps only partly. The press storyline is incidents and safeguards, plus questions about how user chats are handled. The answers most often placed OpenAI among frontier AI developers in safety-index comparisons. Where the data-trust question came up, OpenAI was either absent from the visible text or named by Claude as one of the least trusted. The answers did not draw on the press storylines directly, and this snapshot cannot show why.
Strong on some safety evaluations and commitments but with gaps in disclosure and implementation.
What to watch
A comms team would take three things from this snapshot. First, the responsible-AI question is where OpenAI showed up, and the answers there leaned on third-party safety indexes and corporate reports. Those sources shape the description more than press headlines appear to.
Second, the data-trust and privacy questions are a separate exposure. Other companies carried those answers, and Claude's answer named OpenAI among the least trusted. That gap is worth testing with more benchmark questions on user data before drawing conclusions.
Third, these are single runs from five assistants on five questions. Answers vary between runs, so treat them as a starting point for repeated measurement, not a ranking.