01 The contract
Last updated: 2 August 2026
Answer
Every statistic ProminAI publishes appears on this page with its original source, the period it covers, and the date we last verified it. We do not round figures upward, we do not use statistics older than twelve months without re-verifying them, and we do not cite a number we cannot link to.
02 The figures
Click a column heading to sort. “Last verified by us” is empty until a named member of our team has personally re-checked the figure against the source — publication alone does not earn the tag. Wherever a number appears elsewhere on this site, it is rendered from the same record you see here, so a page and this table can never quietly disagree.
| Figure | What it measures | Source | Period | Last verified by us | Link |
|---|---|---|---|---|---|
| ~800M | weekly ChatGPT users worldwide | OpenAI / TechCrunch | February 2026 | awaiting re-verification | link pending |
| ~100M | weekly ChatGPT users in India | OpenAI / TechCrunch | February 2026 | awaiting re-verification | link pending |
| 1B+ | monthly users of Google AI Mode | Google I/O | May 2026 | awaiting re-verification | link pending |
| ~2B | monthly users of Google AI Overviews | May 2026 | awaiting re-verification | link pending | |
| 60% | of Indian online searchers begin at least one search a day with an AI assistant | YouGov | April–May 2026 | awaiting re-verification | link pending |
| 73% | of Indians use AI to help with shopping and buying decisions | Adobe | being confirmed | awaiting re-verification | link pending |
| 86% | of Indian consumers likely to use gen-AI for at least half of purchase decisions | Accenture Consumer Pulse 2026, India | 2026 | awaiting re-verification | link pending |
| 90% | of Indian consumers would trust an AI agent's recommendation over a best friend's | Accenture Consumer Pulse 2026, India | 2026 | awaiting re-verification | link pending |
| 36% | of Indian consumers would switch brands on an AI recommendation | Accenture Consumer Pulse 2026, India | 2026 | awaiting re-verification | link pending |
| ~65% | of Google searches end without a click | SparkToro / Datos | being confirmed | awaiting re-verification | link pending |
| ≈4.4× | better conversion from AI-referred visitors than ordinary organic search | Semrush | being confirmed | awaiting re-verification | link pending |
| +68% | more time on site from AI-referred visitors | SE Ranking | being confirmed | awaiting re-verification | link pending |
| 94% | of business buyers used a gen-AI tool in their most recent purchase process | 6sense | being confirmed | awaiting re-verification | link pending |
| ≈3× | more likely to be cited by AI systems — brands with 4+ matched sameAs profiles vs 0–1 | Published entity-SEO research | 2026 | awaiting re-verification | link pending |
| ~23% | more AI citations for sites with a well-structured llms.txt than matched controls | Princeton GEO-bench | May 2026 | awaiting re-verification | link pending |
Honest status: 15 of 15 figures are currently awaiting our verification pass — each is published with its named source and period, carries no VERIFIED tag until a person has re-checked it, and will be corrected or removed if the source does not support it. This line disappears only when the count is zero.
03 The rules
A number without its source and time period is an opinion wearing a costume. Every figure we print carries both, in the same breath — on this page, in page copy, everywhere.
If a report says roughly 800 million, we write ~800M — not 1 billion, not “nearly a billion”. Rounding up is the polite name for inflating, and inflation is the industry habit this page exists to oppose.
Five loosely-related numbers in a row feel persuasive and inform nobody. We use the one figure that carries the point and leave the rest here, where they can be checked in context.
AI adoption figures age like fruit, not wine. Anything older than twelve months is re-verified against its source or removed — the helper that flags stale figures runs on this very page.
How many people use ChatGPT every week? Reports from early 2026 do not agree. The internal industry proposal this service grew from cited “900M+ weekly users”. Public reporting around OpenAI's February 2026 India announcement put the global figure at roughly 800 million weekly users, with India around 100 million. Same quarter, same platform, figures a hundred million apart — that is what fast-moving adoption data looks like.
What we publish and why: we use ~800M weekly ChatGPT users worldwide SOURCE: OPENAI / TECHCRUNCH · FEBRUARY 2026 and ~100M weekly ChatGPT users in India SOURCE: OPENAI / TECHCRUNCH · FEBRUARY 2026 — the smaller figures, because they are the ones a named public source supports. When the numbers move and the sourcing is solid, this page moves with them, and the old figure is replaced, not quietly deleted.
We could have printed the bigger number — most of this market does. The fact that we didn't is the fastest way to understand how your own figures will be treated.
04 Cite this page
Link it, quote it, build on it — that is what it is for. And if you spot a figure that has moved, tell us and we will update it — with credit. Corrections make this page more useful, and public corrections are rather the point of it. If you need a figure we have not collected — Indian AI adoption by sector, platform usage by city — ask; if we can source it properly, it joins the table.
Next step
The statistics above describe the market. The free AI Presence Check describes you: what selected platforms return for your real buying questions, recorded verbatim with dates.
Prefer to talk first? Talk to us on WhatsApp — we reply within one working day.
Not ready for a conversation? Start with the ₹3,499 Starter and judge the work first.