Skip to main content

AI Companies · AI Perceptions snapshot

When AI assistants rank AI companies, OpenAI, Google DeepMind and Anthropic carry the story

In this snapshot, the same trio anchors answers on trust, reliability, value and safety, and safety is the one theme the answers credit most often. The press is covering a different set of companies, and a different set of stories, than the assistants are describing.

  • OpenAI
  • Google DeepMind
  • Anthropic
  • Microsoft AI
  • Meta AI
  • xAI
  • Nvidia
  • Mistral AI
  • DeepSeek
  • Perplexity
  • Hugging Face
  • Midjourney
  • Cohere
  • Scale AI
  • Stability AI

Measured September 29, 2026ChatGPT, Claude, Gemini, Perplexity and Grok250 questions · 1,250 answers

Named most in these answers

See the evidence

Measured
September 29, 2026
Assistants
5
Questions
250
Answers
1,250
Brands
15
Named most
OpenAI · 69%95% range 67–72%

Storylines in the answers

The threads the answers kept returning to, and the brands attached to each.

  1. Safety and trust, anchored on the leading labs
  2. Productivity and enterprise integration
  3. Sustainability and ethics
  4. Challengers named far less often than the leaders

What the assistants say

The measurement ran on 2026-09-29. It covered 250 benchmark questions across ChatGPT, Claude, Gemini, Perplexity and Grok, none of which named a brand. That produced 1,250 answers, and 60 of them were read closely for this piece. Across trust, reliability, value, privacy and safety questions, the answers keep returning to OpenAI, Google DeepMind and Anthropic. Anthropic was named in the largest share of trust and reputation questions. OpenAI led on safety and security questions. Google DeepMind led on innovation and capability.

Safety is where the answers disagree in tone. Claude's answer to the safety-commitments question said "Anthropic appears to have the strongest reputation". Grok's answer to the same question said no major company has a flawless record. Perplexity's answer on incident response said public evidence was too thin to rank the labs. The trio is named, but many answers attach a caveat to the naming.

Microsoft AI and Google DeepMind carry a second storyline, productivity and enterprise integration. ChatGPT's answer to the "forward-thinking about productivity software" question named Microsoft as the probable leader. Microsoft AI also placed second among brands named in sustainability and ethics questions.

Anthropic appears to have the strongest reputation…
Claude, answering “Which AI companies are known for keeping their promises on safety commitments?”

Who owns the answer, and why

OpenAI was named in 694 of 1,000 answers, Google DeepMind in 691 and Anthropic in 662. Each provider's own brand is excluded from its own answers, which is why those totals are out of 1,000. Microsoft AI appeared in 512 of 1,250 answers and Meta AI in 426. Below them the field thins out: Nvidia in 151, Mistral AI in 149, DeepSeek in 135, xAI in 133 and Perplexity in 85.

The credit attached to those names is thin. "Not Substantive" is the largest tag for every major brand, meaning the name appears without a specific reason attached. Where a reason is tagged, Safety & Compliance leads for OpenAI, Meta AI, Anthropic and Google DeepMind, and Sustainability & Ethics leads for Microsoft AI. The tag counts are small next to the naming counts.

The cited domains point the same way. openai.com, anthropic.com and help.openai.com are the most cited domains, and owned sources make up 58% of the source mix in this snapshot. News is 10% and trade press 8%. No Google-owned domain appears among the eight most cited. The Future of Life Institute site (futureoflife.org) is in that top eight, consistent with the safety-index material some answers drew on.

No major AI companies have a flawless or strongly consistent record of fully upholding ambitious safety commitments, according to independent assessments and reporting.
Grok, answering “Which AI companies are known for keeping their promises on safety commitments?”

What the press is covering

In the PeakMetrics news index over the past 30 days, OpenAI had about 156,000 mentions, Nvidia about 150,000 and Anthropic about 133,000. Google DeepMind had about 44,000, Microsoft AI about 41,000 and Hugging Face about 39,000. Mistral AI had about 1,100 and Stability AI about 330.

The headlines differ by brand. Anthropic's recent headlines concern a disclosed hacking incident, a researcher quitting over safety, and its CEO urging the industry to slow model development. OpenAI's combine a security-incident investigation, a GPT-6 launch and a dispute over a mathematics result. Google DeepMind's coverage includes reports of Gemini hacking three companies during testing. Meta AI's headlines are about a proposed class action and smart-glasses tampering. Microsoft AI's are mostly investor and product-webinar stories. Nvidia's are about a share buyback and its CEO rejecting doomsday fears. DeepSeek's are about a price war and a briefing to the UN Security Council.

There isn’t enough public, comparable evidence to confidently rank OpenAI, Anthropic, Google DeepMind, and other labs.
Perplexity, answering “What AI companies have the best incident response when their systems misbehave?”

Where the two stories diverge

The widest gaps are in coverage volume. Nvidia drew about 150,000 news mentions but was named in 151 of 1,250 answers. Hugging Face drew about 39,000 and Scale AI about 22,000, yet they appeared in 83 and 32 answers. Mistral AI and Stability AI are the reverse: their news coverage is small, and each still appears in roughly the same number of answers as far more heavily covered peers (149 and 32). In this snapshot, press volume and assistant visibility look separate.

There is also a gap in subject matter. The safety answers lean on safety indexes, frameworks and commitments, and several name Anthropic as the reference point. The recent Anthropic headlines are about incidents and internal dissent over safety. Both are attributed descriptions of different source material, not a verdict, but a reader who saw only one would miss the other. Hugging Face shows the pattern on a smaller scale. Its recent headlines concern a breach and an acquisition, while in the answers it surfaced mainly as a place to find benchmarks.

What this means for comms teams

Companies outside the trio are named in a small share of answers, and the tags rarely say why. The basics for those companies are the pages the assistants can cite: clear owned pages on safety, privacy, pricing and support. The most cited domains in this snapshot are owned pages, and help-center content is among them.

Third-party evaluations shape the safety answers. A safety index was a recurring source, and the answers quote its gradings. Companies should know what outside indexes say about them and whether the facts those indexes rely on are current.

Press volume is not a proxy for assistant visibility. Teams should track both. This is one snapshot from live answers that vary between runs, so treat the figures as a baseline to re-measure, not a ranking.

The evidence

Which brands the answers named, and what they cited.

ChatGPT, Claude, Gemini, Perplexity and Grok answered 250 unbranded questions about this sector, 1,250 answers in all.

Visibility leaderboard

Ordered by visibility: the share of this sector's answers that named each brand. Each figure comes with a 95% range; where two brands’ ranges overlap, they are effectively tied. This is not a favorability ranking.

Visibility by brand

  1. OpenAI
    69%67–72%
  2. Google DeepMind
    69%66–72%
  3. Anthropic
    66%63–69%
  4. Microsoft AI
    41%38–44%
  5. Meta AI
    34%32–37%
  6. xAI
    13%11–16%
  7. Nvidia
    12%10–14%
  8. Mistral AI
    12%10–14%
  9. DeepSeek
    11%9–13%
  10. Perplexity
    9%7–10%
  11. Hugging Face
    7%5–8%
  12. Midjourney
    3%3–5%
  13. Cohere
    3%2–4%
  14. Scale AI
    3%2–4%
  15. Stability AI
    3%2–4%
Measured September 29, 2026. ChatGPT, Claude, Gemini, Perplexity and Grok, 1,250 answers. The line through each bar is its 95% range, printed under the figure. A brand’s own assistant is not counted in its figures.
Measured September 29, 2026. ChatGPT, Claude, Gemini, Perplexity and Grok, 1,250 answers. “Own” marks a brand’s own assistant, whose answers are excluded from its figures.
RankBrandVisibilityChatGPTClaudeGeminiPerplexityGrokNamed in
1OpenAI
69%95% range 67–72%
own assistant, excluded59%70%57%92%694 / 1,000
2Google DeepMind
69%95% range 66–72%
75%56%own assistant, excluded54%92%691 / 1,000
3Anthropic
66%95% range 63–69%
67%own assistant, excluded66%45%87%662 / 1,000
4Microsoft AI
41%95% range 38–44%
48%30%45%29%53%512 / 1,250
5Meta AI
34%95% range 32–37%
28%24%34%25%58%426 / 1,250
6xAI
13%95% range 11–16%
16%11%12%14%own assistant, excluded133 / 1,000
7Nvidia
12%95% range 10–14%
9%10%11%8%22%151 / 1,250
8Mistral AI
12%95% range 10–14%
10%7%13%8%22%149 / 1,250
9DeepSeek
11%95% range 9–13%
3%9%7%9%26%135 / 1,250
10Perplexity
9%95% range 7–10%
5%6%8%own assistant, excluded15%85 / 1,000
11Hugging Face
7%95% range 5–8%
4%4%7%2%15%83 / 1,250
12Midjourney
3%95% range 3–5%
2%2%4%3%6%42 / 1,250
13Cohere
3%95% range 2–4%
5%2%1%2%5%39 / 1,250
14Scale AI
3%95% range 2–4%
2%2%2%2%5%32 / 1,250
15Stability AI
3%95% range 2–4%
2%3%3%2%3%32 / 1,250

Top cited domains

The domains the answers cited most often, with each one's share of all citations in this sector. These are the most-cited sources, not every source cited. Bars are scaled to the most-cited domain.

  1. openai.com
    142 citations2%
  2. anthropic.com
    101 citations1%
  3. arxiv.org
    74 citations1%
  4. help.openai.com
    70 citations1%
  5. medium.com
    69 citations1%
  6. futureoflife.org
    54 citations1%
  7. forbes.com
    51 citations1%
  8. techcrunch.com
    44 citations1%

Source type breakdown

The type of site behind this sector's 8 most-cited domains, which account for 605 of the 7,560 citations in these answers. It is not the mix of every citation. Shares are of the classified citations only; domains the classifier could not label confidently are left out. Labels come from a classifier run on this site, not from a customer workspace's Smart Categories.

  • Owned313 citations · 58%
  • Other128 citations · 24%
  • News51 citations · 10%
  • Trade44 citations · 8%

Questions asked

250 unbranded questions across 13 intents, each put to every assistant once. None names a brand, so a brand that appears in an answer was volunteered by the model. The 5 seed questions, marked below, are the ones earlier five-question snapshots asked.

  • Reliability and quality21 questions
    1. Which AI models are the most reliable for business use?Seed
    2. Which AI chatbot makes the fewest factual mistakes?
    3. I run a small customer support team and need an AI model that gives consistent answers day after day. Which providers should I look at?
    4. Which AI companies have the worst outages?
    5. Are there AI providers that don't suddenly change their models and break my workflow?
    6. What's the most consistent AI image generator for professional work?
    7. I'm a lawyer worried about made-up citations. Which AI tools are least likely to invent sources?
    8. Which AI providers publish honest status pages and reliability records?
    9. Who has the most polished AI product overall, with the fewest glitches and bugs?
    10. Are open-source model makers as dependable as the big commercial AI labs?
    11. Which AI companies consistently ship updates that make their models better rather than worse?
    12. What are the most reliable AI translation tools for business documents?
    13. Which AI chatbot slows down or gets rate-limited the least during peak hours?
    14. As a teacher, which AI writing assistants give the most dependable results for grading help?
    15. Which AI companies have a track record of delivering on their product promises?
    16. Which enterprise AI platforms have the best service-level guarantees?
    17. Who makes the most reliable AI search assistant for current events?
    18. Which AI coding assistant would you trust most for a production codebase?
    19. Do any AI model makers have a reputation for their models getting worse over time?
    20. I'm a journalist who needs an AI research tool that cites real sources correctly. Which company does this best?
    21. Which AI chatbots are known for making up false information about real people?
  • Trust and reputation11 questions
    1. Which AI companies are the most trustworthy?Seed
    2. Who are the most honest AI labs when it comes to admitting their models' limitations?
    3. What's the worst-reputation AI company right now?
    4. As an investor, which AI companies have the strongest reputation for governance and honest leadership?
    5. What AI companies have the best reputation among journalists for being open and responsive?
    6. I'm 68 and nervous about AI. Which companies are the most reputable and least likely to mislead me?
    7. Which AI companies have had the most public trust problems over the past two years?
    8. Which big AI developers do the public trust more than the tech industry overall?
    9. What are the most trusted AI providers for government agencies?
    10. Are open-source AI developers generally more trusted than closed-source ones, and which are the best regarded?
    11. Should I trust the AI companies that say they're building safe AI, and which ones have earned it?
  • Value and price23 questions
    1. Which AI assistant offers the best value?Seed
    2. Is paying for a premium AI chatbot subscription actually worth it compared to the free versions?
    3. What's the cheapest AI assistant that still handles coding well?
    4. I'm a college student on a tight budget and need an AI tool for writing and research. Which paid or free plan gives me the most for my money?
    5. How do the monthly prices of the major AI chatbots compare right now?
    6. Are the $200-a-month AI plans worth it for anyone other than power users?
    7. Which AI company has the most generous free tier?
    8. I run a five-person marketing agency and we're deciding between team plans from different AI providers. Which one gives the best bang for the buck?
    9. Are there any AI image generators that are actually worth paying for, or are the free ones good enough?
    10. What are the hidden costs of using enterprise AI platforms that companies often overlook?
    11. Which AI provider offers the best discounts for nonprofits or educators?
    12. Do open-source AI models save money compared to paying for a proprietary one, once you count hosting costs?
    13. Which AI coding assistant is the best deal for a freelance developer?
    14. My small law firm wants an AI tool for document review, and I'm worried about ballooning costs. Which vendors have the most predictable pricing?
    15. Which AI companies charge extra for things like image generation, voice mode, or longer context limits?
    16. Is it worth paying for a premium AI plan just to get higher usage limits?
    17. What's the best value AI voice generator for making podcasts or videos?
    18. Who has the cheapest AI video generation tools that still look decent?
    19. Are AI company price cuts making it smarter to wait before signing an annual plan?
    20. I'm retired and just want an AI helper for emails, recipes, and travel planning. Do I need to pay for one, and which is the best deal if so?
    21. Which AI provider gives the best performance per dollar for customer support chatbots?
    22. Are the cheaper, smaller AI models from the big labs good enough for everyday tasks, or is the flagship model worth the extra cost?
    23. How much do the top AI companies charge for enterprise licenses per seat?
  • Customer service21 questions
    1. Which AI company has the best customer support when something goes wrong?
    2. Which AI subscriptions let you cancel easily without getting locked into a yearly contract?
    3. I got charged twice for my monthly AI subscription and nobody is answering my emails. Which AI companies are actually good about issuing refunds?
    4. Do any AI companies have real human support, or is it all bots answering your tickets?
    5. Who has the worst customer service among the big AI model makers?
    6. My AI account got suspended for no clear reason and I can't get anyone to explain why. Which providers have a decent appeals process?
    7. What AI companies offer phone support for paid subscribers?
    8. Are there AI providers that give a prorated refund if you cancel an annual plan early?
    9. How do the top AI companies handle complaints when the chatbot gives harmful or wrong advice?
    10. best AI company for enterprise support?
    11. Which AI image generator companies have the fairest refund policies when the tool fails to deliver what you paid for?
    12. Which AI vendors have the most responsive support teams for developers hitting rate-limit or usage-cap problems?
    13. What do people say about AI companies that ban accounts automatically and then ignore appeals?
    14. Is there an AI company that actually offers a money-back guarantee on its paid plans?
    15. I'm a first-time buyer of an AI writing tool subscription. Which companies are easiest to get help from if I run into trouble?
    16. Which AI companies have the best help centers and documentation for solving problems on your own?
    17. How do AI companies compare on how quickly they reply to support tickets?
    18. Which AI model makers are known for sneaky auto-renewal or hidden billing practices?
    19. Which AI providers offer dedicated account managers for mid-size business customers?
    20. What are the most common customer complaints about the major AI chatbot companies, and which ones handle them best?
    21. Where can I file a complaint about an AI company, and which ones tend to actually respond?
  • Safety and security22 questions
    1. Which AI companies are most responsible about safety?Seed
    2. Which AI companies are most respected by independent safety researchers?
    3. Are there AI companies that publish real safety reports instead of just marketing claims?
    4. Which AI companies are known for keeping their promises on safety commitments?
    5. If an AI company leaks my chat history, which ones have a track record of handling it well and notifying customers?
    6. Has any major AI company had a serious data breach?
    7. How do the big AI labs compare on red-teaming and pre-release safety testing?
    8. What's the safest AI chatbot for my 10-year-old to use for homework?
    9. Which AI companies have the worst track record on harmful or biased outputs?
    10. What AI companies have had to pull back or roll back a model release because of safety problems?
    11. Which AI assistants are best at refusing to give dangerous instructions?
    12. Who has the best bug bounty and vulnerability disclosure program among AI companies?
    13. Which AI company is best at preventing jailbreaks and prompt injection attacks?
    14. Which AI image generators have the best safeguards against deepfakes and abusive content?
    15. Which AI companies have dedicated safety teams, and which have cut them back?
    16. What are the most secure AI options for government agencies or defense contractors?
    17. Which open-source AI model developers do the most to prevent misuse?
    18. Which AI companies do independent third-party safety audits?
    19. Which AI chatbots are safest for people dealing with mental health struggles?
    20. What AI companies have the best incident response when their systems misbehave?
    21. How do AI companies compare on transparency about their safety testing and their societal impact?
    22. I run a small law firm and want to use AI tools. Which providers have had confidentiality or data leak problems I should know about?
  • Innovation20 questions
    1. Which AI companies are leading the way in new model breakthroughs right now?
    2. Who's actually innovating in AI and who's just copying everyone else?
    3. I'm a software engineer picking an AI coding assistant for my team. Which companies are pushing the most innovative features in that area?
    4. Which AI companies are making the biggest advances in video generation?
    5. Who has the most impressive voice assistant among the big AI model makers?
    6. Rank the top AI companies by how fast they ship new products.
    7. Are there any smaller AI startups out-innovating the big labs?
    8. My company wants to modernize customer service with AI. Which vendors offer the most advanced technology for that?
    9. Which AI company has the best app experience on phones?
    10. Which AI makers are most innovative at making models smaller and more efficient?
    11. Who is ahead in AI for scientific research and drug discovery?
    12. As an investor, which AI model makers look the most technologically ahead of the pack?
    13. What's the most user-friendly AI chatbot interface, and which company does it best?
    14. Which AI company is the most forward-thinking about integrating AI into everyday productivity software?
    15. Are there AI companies making real progress on robots and physical-world AI?
    16. Which AI labs are leading in custom chip and hardware development for their models?
    17. I'm a teacher looking for AI tools that feel modern and innovative for the classroom. Which companies are doing the best work here?
    18. Who has the best AI search experience compared to traditional search engines?
    19. Which AI companies have released the most useful new features in the last year?
    20. Which model makers are pioneering new ways to personalize AI assistants with memory?
  • Leadership and scale21 questions
    1. Who are the biggest AI companies in the world right now?
    2. Which AI company is winning the race for market share?
    3. Which AI startups have raised the most venture funding?
    4. Who are the top five AI model makers ranked by overall influence?
    5. Which AI company has the strongest balance sheet?
    6. Which AI labs are best positioned to still be around in ten years?
    7. I'm an investor trying to understand the AI landscape. Which companies are the clear market leaders in foundation models?
    8. Who is leading in AI right now, the big tech giants or the startups?
    9. Which AI company is the best run?
    10. Which AI companies are actually profitable?
    11. Top AI companies by valuation?
    12. Which AI company has the most users worldwide?
    13. Which AI companies have the most influence on government policy in Washington?
    14. Rank the leading American AI labs by size and resources.
    15. Which AI company spends the most on computing infrastructure and data centers?
    16. Are there any AI model makers outside the US that are serious competitors to the leaders?
    17. Which AI companies have the best-known research teams?
    18. I'm writing an article on who really controls the AI industry. Which companies should I be focusing on?
    19. Which open-source AI model makers have the biggest impact on the field?
    20. Which AI companies have the strongest partnerships with big cloud providers?
    21. Who are the dominant players in enterprise AI sales?
  • Employer and workplace21 questions
    1. Is there an AI company that's genuinely well regarded by its own former employees?
    2. If I'm a job seeker who cares about ethics, which AI companies have the best reputations as employers?
    3. As a college grad hoping to work in AI, which companies are the most prestigious and stable to join?
    4. What are the best AI companies to work for right now?
    5. I'm a new grad with a machine learning degree. Which AI labs offer the best starting salaries and training?
    6. Which AI companies have the best work-life balance?
    7. Are AI startups worth the risk compared to working at a big tech company's AI division?
    8. How much do research scientists make at the top AI labs?
    9. Which AI companies have the worst reputation among employees?
    10. Best AI employers for someone who cares about safety research?
    11. Do any AI model makers have unions or active labor organizing?
    12. I'm 45 and thinking about switching from software engineering into AI. Which companies are open to career changers?
    13. Which AI companies offer the best equity packages for employees?
    14. What's the culture like at the leading AI labs, and which ones are known for burnout?
    15. Who pays the most for prompt engineers?
    16. Are there AI companies that let employees work fully remote?
    17. Which AI labs have had the most employee departures or walkouts over ethics concerns?
    18. Which AI companies treat their data labelers and contractors the best?
    19. Which AI companies are most diverse and inclusive for women engineers?
    20. Best AI companies for non-technical careers like policy, marketing, or operations?
    21. Have any big AI companies done layoffs recently, and which ones seem most stable for job security?
  • Sustainability and ethics21 questions
    1. Do any AI companies have a good track record on being upfront about energy use and emissions?
    2. Who do AI ethics experts generally consider the good guys in the industry?
    3. Which AI firms are seen as the most credible on responsible AI, and which are just talking?
    4. Which AI company has the most consistent reputation for ethical behavior?
    5. Which AI companies are doing the most to cut the carbon footprint of training their models?
    6. How do the big AI labs compare on water use at their data centers?
    7. Which AI developers have the strongest independent oversight or board governance?
    8. What AI companies have committed to running their data centers on renewable energy?
    9. I'm an investor who cares about ESG. Which publicly traded AI companies score best on sustainability?
    10. Best AI company for someone who wants to avoid contributing to local power grid strain?
    11. Which AI companies have the worst environmental record?
    12. My town is being pitched a big AI data center. Which AI companies have a good track record of working with the communities around their facilities?
    13. Which AI companies are structured as public benefit corporations or nonprofits, and does that make them more accountable?
    14. Which AI companies are the most open about the biases in their models?
    15. What are the most eco-friendly AI tools for a small business trying to shrink its carbon footprint?
    16. Are open-source AI model makers more ethical than closed ones?
    17. Rank the major AI companies by how seriously they take social responsibility.
    18. Which AI providers offer carbon-neutral or low-emissions cloud options for running models?
    19. Which AI companies have the best policies on preventing their tech from being used for surveillance or weapons?
    20. Do any AI companies contribute meaningfully to communities through education, grants, or local job programs?
    21. Which AI companies have independent audits of their environmental claims?
  • Avoid and risk14 questions
    1. Which AI company should businesses avoid?Seed
    2. As an investor, which AI companies look most exposed to safety or security liabilities?
    3. As a journalist covering AI, which companies should I look at for the most troubling governance practices?
    4. What are the biggest red flags when picking an AI vendor for my company?
    5. Are there AI startups that turned out to be overhyped or misleading investors?
    6. I'm thinking of investing in AI stocks. Which companies carry the most legal or regulatory risk?
    7. As a journalist, which AI firms have faced the most scandals over the past couple of years?
    8. What AI companies have faced regulatory investigations in the US or Europe?
    9. Should I worry about using AI coding assistants from certain companies because of licensing or security risks?
    10. Are any AI companies accused of exploiting low-paid data labelers overseas?
    11. Which AI chatbots have been banned or restricted by governments or schools?
    12. Top AI companies with the most controversies?
    13. I work in healthcare. Which AI providers should I be cautious about because of privacy or accuracy issues?
    14. Which AI firms have been accused of misleading marketing or overstating what their products can do?
  • Data privacy and training data21 questions
    1. Which AI company has the best reputation for protecting user privacy?
    2. I'm a lawyer and worried about client confidentiality. Which AI chatbot makers can I actually trust with sensitive documents?
    3. Are there any AI companies that have never been caught misusing customer data?
    4. Which AI developers are considered the most transparent about how they train their models?
    5. My teenager uses an AI chatbot for homework. Which companies are actually trustworthy with kids' data?
    6. Which AI company would you trust least with your personal information?
    7. Which AI labs have been most criticized for scraping copyrighted work, and which ones license their training data?
    8. I'm a hospital IT director. Which AI vendors have the best reputation for handling health data responsibly?
    9. Which AI chatbot makers are the most upfront about not using your conversations for training?
    10. Which AI companies get the best marks from watchdog groups on data practices?
    11. Which AI companies make it hardest to delete your account and data?
    12. Which AI companies let you opt out of having your conversations used to train their models?
    13. Are AI companies that store chat history less safe than ones that delete it quickly?
    14. My company is worried about employees pasting trade secrets into AI tools. Which vendors offer the best enterprise data controls?
    15. Is it safe to upload tax documents or medical records to an AI chatbot, and which ones handle that data best?
    16. Which AI companies have faced regulatory fines or investigations over privacy?
    17. How do AI companies compare on protecting user data from being sold to advertisers?
    18. Best AI coding assistant that doesn't send my proprietary code to outside servers?
    19. Are there AI companies that let artists and writers opt out of having their work used for training?
    20. Is there an AI chatbot made by a company with a genuinely good privacy and ethics record?
    21. Which AI image generators have the worst reputation for copying artists' work?
  • Model capability and performance21 questions
    1. Who makes the best AI model for writing code without lots of bugs?
    2. Which AI models are best at following long, detailed instructions without drifting?
    3. best AI transcription service for accuracy?
    4. Which AI assistant is the best at math and logic without making arithmetic slips?
    5. What AI company has the highest quality voice assistant that actually understands people?
    6. Which AI tools are best for summarizing long documents accurately?
    7. Which AI lab has the most advanced reasoning models today?
    8. Which AI company has the best multimodal model that handles text, images, audio, and video together?
    9. Who leads in AI image generation quality right now?
    10. Which AI companies are doing the most interesting work on long-context models that can read entire books or codebases?
    11. What AI company has the best real-time translation and speech technology?
    12. Who has the most powerful AI models at the moment?
    13. What's the best AI for solving hard math problems, like college-level calculus and proofs?
    14. I'm a software engineer with a big legacy codebase. Which AI models handle long context well enough to review thousands of lines at once?
    15. Which AI image generator produces the most realistic photos?
    16. Are there AI models that can actually reason through multi-step logic puzzles without making mistakes?
    17. How do the top AI models compare on the major benchmarks?
    18. Which AI has the largest context window?
    19. What's the best AI for creative writing, like fiction with a consistent voice?
    20. Which AI is best at understanding images, charts and screenshots?
    21. Is there an AI that can reliably turn a hand-drawn sketch into working front-end code?
  • Developer and enterprise integration13 questions
    1. Are there any AI model makers that offer on-premise or private cloud deployment for maximum security?
    2. What AI companies are leading in open-source models that developers can run themselves?
    3. Which AI providers have the most innovative tools for building autonomous agents?
    4. Top AI companies for enterprise innovation and modern cloud integration?
    5. What's the best AI provider for building a retrieval-augmented generation system on top of our internal documents?
    6. Which open-source AI models are best for commercial use without restrictive licenses?
    7. Which cloud platforms offer the widest selection of hosted foundation models for enterprise teams?
    8. What are the best AI coding assistants to integrate into an enterprise engineering workflow?
    9. Which AI vendors offer dedicated capacity or provisioned throughput for large enterprise customers?
    10. Which providers have the best embeddings models for semantic search?
    11. What are the best small language models for running on edge devices or mobile apps?
    12. I run a mid-size healthcare software company and want to add generative AI features. Which providers should be on my shortlist?
    13. Which AI companies have the best developer community and ecosystem of third-party libraries?

How this was measured

Measured September 29, 2026. 250 unbranded questions per sector, asked to ChatGPT, Claude, Gemini, Perplexity and Grok on their product models, with web search on. Refreshed quarterly.

This is a snapshot, not a ranking or a trend: AI answers vary from run to run. It shows whether a brand appeared and which sources were cited, not how favorably it was described. The paid platform tracks favorability daily.

An assistant naming its own maker is not counted: ChatGPT's answers are left out of OpenAI's figures, Gemini's answers are left out of Google DeepMind's figures, Claude's answers are left out of Anthropic's figures, Grok's answers are left out of xAI's figures and Perplexity's answers are left out of Perplexity's figures.

Full method

In an early five-question test, the same brand went from 40% to 30% to 70% visibility across three runs, which is why every figure here is dated and treated as a snapshot.

The questions cover 13 intents, such as reliability, value for money and customer service. A model drafted them, a review pass removed paraphrases and questions that point to one company, and every question was checked so that none names a measured brand. Each was asked once per assistant in run 2026Q4.

Every visibility figure comes with a 95% range (a Wilson interval). Where two brands' ranges overlap, treat them as tied. The figures use all 1,250 answers. The quotes in the story come from a stratified sample of 60 of them, spread across every assistant and intent.

Summarized on September 30, 2026from this snapshot's stored answers. Every quote is checked against the original answer.

← All sectors

Other sectors

See how AI answers describe your brand every day.

AI Perceptions tracks all 5 providers daily, with the cited sources behind each answer and the news coverage around it.

Book a Demo