AI assistant usage is rising fast. Trust is not. Kantar BrandDynamics shows Claude was the only major AI brand to post a statistically significant Demand Power gain in the most recent 30 day read, and the attributes that produced it had been building since January. Two decades of BrandZ evidence say puts a price on the usage-trust gap.
The most-adopted, least-trusted technology in modern memory
The AI assistant category is living a paradox no consumer technology has faced at this scale. Adoption is compounding with 44% of U.S. adults now reporting they use ChatGPT, up from 18% in 2023. Trust has not moved with it. Only 29% of chatbot users say they have even some trust in the information chatbots provide, 71% of U.S. adults believe AI advances will weaken personal data security, and 52% say they are more concerned than excited about AI in daily life, up from 37% in 2021. That last figure has been reinforced in recent weeks by several AI company leaders publicly warning about the technology's risks.
For the companies building these products, the gap between usage and belief is the single largest unpriced risk in the category. When a market grows this fast, distribution and model quality decide who gets tried. Brand decides who gets kept, who gets paid a premium, who wins the enterprise procurement committee, and who survives the first genuinely bad news cycle.
Many of these companies already run brand trackers as a lagging indicator. The harder question is what those trackers are connected to. Brand health read as a marketing scorecard tells you how paid, earned and owned media performed. Brand health read as a leading indicator tells you where demand is heading two quarters out, and what the brand will be worth when it arrives. Two decades of Kantar BrandZ evidence make the second reading the more valuable one.
What brand health predicts about enterprise value
Kantar BrandZ has valued the world's brands for more than twenty years, and the pattern in that data is consistent enough to be uncomfortable. A portfolio of strong brands returned 435% from April 2006 through March 2025. The S&P 500 returned 353% over the same window, and the MSCI World Index returned 171%. The gap widens most during downturns, which is when a brand stops behaving like a marketing asset and starts behaving like a financial shield.
The mechanism underneath that return is measurable at the brand level. Brands with high equity take roughly nine times the volume share of brands with low equity, a relationship validated by the Marketing Accountability Standards Board using the same Meaningful, Different and Salient framework that Kantar BrandDynamics reports continuously.
In technology specifically, that link showed up again this year. All ten of the most valuable consumer technology and services brands grew brand value between 2025 and 2026, together worth $1.9 trillion. Every one of them did it by improving Meaningful Difference, Salience, or both, and the largest value increases belong to the brands that improved both. Value did not rise where those measures did not move first.
| Brand | 2026 brand value | Change vs 2025 | What moved |
|---|---|---|---|
| Nvidia | $814.9B | +60% | US brand equity among business buyers, up 8x since 2020 |
| ChatGPT | $167.8B | +285% | Year's number one riser; number one in Demand Power among US consumers |
| IBM | $150.8B | +20% | Brand Trust index of 129, built over decades |
| Claude | $96.6B | New entry | Enterprise adoption; US Future Power at 111 |
Table 1. Kantar BrandZ 2026, Consumer Technology and Services and Business Technology and Services rankings, with US and global brand equity series.
None of this is new to platform categories. At every prior inflection in search, social, mobile and streaming, consumer predisposition formed early, hardened fast, and then taxed or subsidized everything the company did afterwards: acquisition costs, pricing power, enterprise sales cycles, talent, regulatory goodwill. Instagram's brand value grew 1,479% between 2018 and 2025 against 144% for the BrandZ Global Top 100, and that gap was built during the years the category's defaults were being set, not after they had settled.
That window is open again right now. Meta's Muse, a personal AI agent the company recently launched and has been pushing hard since, is exactly the kind of entrant whose defaults are still being written. We're adding Muse to the BrandDynamics tracker and will have a first read on where it lands soon.
The forward read: Future Power
Retrospective evidence only takes the argument so far. BrandZ also carries a forward measure, Future Power, which estimates a brand's growth potential relative to its current size. A score of 100 is par. Above it, a brand is carrying more predisposition than its present share would suggest, and that surplus has a habit of converting.
The clearest demonstration comes from a category nobody would have called. In 2020, US Future Power put DoorDash at 120, against UberEats at 101 and GrubHub at 92. Six years later DoorDash entered the BrandZ Consumer Technology ranking for the first time, valued at $26.5 billion. The measure identified the outcome while the market still looked like a three-way fight.
The same read is already available for AI brands. Claude's Future Power stands at 111 in the US, which is why its entry into the Top 100 is better understood as a starting position than a peak. Future Power is the bridge between the attribute movements and the valuation. The attributes move first, Future Power reads what they imply, and brand value follows.
Why AI raises the stakes further
AI is at that moment now, with an aggravating factor no prior shift carried. AI already has a trust problem: 71% of the public expects it to put their data security at risk. Trust here is the license to operate. The brands accumulating it now are buying cheap insurance against the category's first major crisis, whichever brand triggers it. Brands riding salience alone may find that fame does not hold up under pressure.
Which raises the practical question. If brand health predicts future value, which parts of brand health move first?
What continuous brand measurement reveals that usage metrics cannot
Using BrandDynamics to track the U.S. AI assistants category continuously, decomposing each brand's Demand Power, two structural facts jump out of our read of the 30 days ending 27 August 2026.
First, demand is brutally concentrated. ChatGPT holds 14.5% Demand Power and Gemini 11.6%, and no other brand clears 8.5%. This is a winner-take-most structure in which small equity shifts compound into share.
Second, Difference rather than Salience is doing the heavy lifting. In this category Difference contributes 24.3% of Demand Power against a 17.1% market average, while Salience contributes 36.5% against a 45.5% average. Being known is table stakes. Being known for something specific is what converts. That inverts the instinct of most AI marketing teams, whose budgets are built as though awareness were the bottleneck.
Among 21 tracked brands, only Claude posted a statistically significant Demand Power gain in the period.
| Brand | Meaningful | Different | Salient | Demand Power | 30-day chg |
|---|---|---|---|---|---|
| ChatGPT | 160 | 163 | 203 | 14.5% | +0.0 |
| Gemini | 156 | 119 | 188 | 11.6% | +0.2 |
| Google Assistant | 126 | 99 | 165 | 8.4% | −0.1 |
| Alexa+ | 113 | 118 | 144 | 7.3% | −0.1 |
| Microsoft Copilot | 113 | 109 | 143 | 6.9% | −0.1 |
| Meta AI | 96 | 112 | 145 | 6.4% | −0.1 |
| Siri | 106 | 107 | 133 | 6.4% | −0.1 |
| Grok | 91 | 106 | 93 | 3.2% | +0.1 |
| Apple Intelligence | 92 | 105 | 93 | 3.1% | −0.1 |
| Claude | 93 | 104 | 88 | 2.8% | +0.3 ↑ |
| Perplexity AI | 88 | 94 | 81 | 2.3% | −0.0 |
| DeepSeek | 91 | 97 | 75 | 2.3% | +0.0 |
Table 2. Kantar BrandDynamics Demand Power, U.S. AI Assistants (top brands), 30 days ending 27 August 2026. Indices are shown against a category average of 100, so Claude's Salient index of 88 sits below average even as it moved upward in the period. ↑ marks a statistically significant change against three months prior. Sample n=201–568 per brand.
The fame–trust gap: reading the leading indicators
Demand Power tells you the score. The attributes underneath it tell you what the score will be next quarter. Kantar's year-to-date attribute trends for this category, tracked daily through True Signal among category-aware consumers, separate three kinds of brand momentum that look identical in a share-of-voice report.
Borrowed fame
Influencer and celebrity endorsement is the fastest-rising attribute in the category, and every tracked brand gained on it this year, evidence of how thoroughly personalities now mediate AI's public narrative. Grok leads with +4.1 points year to date, rising from 4.8% to 8.9%, an Elon Musk dividend, with Joe Rogan's platform doing similar work for Perplexity. Google Assistant and Alexa+ show the pattern more starkly still, with endorsement rising while trust falls. Fame borrowed from a personality, unaccompanied by earned meaning, is coincident noise rather than a leading indicator.
Earned trust
A second group grew trust faster than fame. DeepSeek, Perplexity and Claude all did, with DeepSeek leading the way, and Claude posting the second-largest privacy gain, from 9.4% to 15.7%.
Habituation
The quietest attribute is the most predictive: fits well in everyday life. Claude and DeepSeek posted the category's largest gains here. Microsoft Copilot fell 3.9 points and Galaxy AI fell 5.3, the steepest decline in the set, though Galaxy AI sits outside the top brands shown in the table below.
| Brand | Endorsement | Trustworthy | Privacy | Everyday fit | DP chg (30d) |
|---|---|---|---|---|---|
| Grok | +4.1 | +1.1 | +1.2 | +3.9 | +0.1 |
| Google Assistant | +3.2 | −3.0 | −0.6 | −0.5 | −0.1 |
| Gemini | +3.1 | +1.7 | +2.6 | +4.4 | +0.2 |
| Alexa+ | +2.8 | −2.4 | −0.6 | −1.9 | −0.1 |
| Meta AI | +2.5 | −2.2 | −1.8 | −1.6 | −0.1 |
| Microsoft Copilot | +2.3 | +0.6 | +0.8 | −3.9 | −0.1 |
| DeepSeek | +2.2 | +5.1 | +8.3 | +7.8 | +0.0 |
| ChatGPT | +1.8 | +1.4 | +4.2 | +1.3 | +0.0 |
| Claude | +1.6 | +2.5 | +6.3 | +7.7 | +0.3 ↑ |
| Perplexity AI | +1.4 | +4.0 | +0.8 | +2.3 | −0.0 |
Table 3. Year-to-date change in percentage points in brand-attribute association among brand-aware U.S. consumers, January to August 2026, Kantar BrandDynamics with True Signal (n=318–529 per brand), alongside 30-day Demand Power change. Brands where trust growth outpaces endorsement growth are building. Brands where fame outruns trust are borrowing.
Now line those movements up against outcomes. Claude's Demand Power gain came through a +6 move in Salience, +1 in Meaningful and +3 in Different. Read on its own, that looks like a fame story. Read against the year underneath it, it is the opposite: everyday-life fit +7.7, privacy +6.3 and trust +2.5, each compounding since January. The salience that surfaced in the 30-day read was the consequence of those earned gains, not a substitute for them, and the attributes recorded it months before Demand Power did.
The reverse held too. The legacy assistants' eroding trust and fit scores appeared months ahead of their synchronized Demand Power decline. By the time demand moved, the causes were already old news.
The leading-indicator dashboard every AI brand should run
The practical agenda that falls out of this data is specific: track Demand Power continuously rather than in annual dips. In a category moving this fast, a yearly brand study isn't strategy, it’s archaeology.
To build a tracker that actually predicts performance, structure your dashboard around five core principles:
- Break Demand Power down into Meaningful, Different, and Salient: This category rewards Difference at 1.4 times the market norm and actively punishes undifferentiated fame.
- Run four core attributes underneath demand
- Trustworthy & Protects my privacy: The non-negotiable license-to-operate pair.
- Fits well in everyday life: The habituation signal that precedes demand shifts by months.
- Influencer and celebrity endorsement: The early-warning tripwire. A fame spike is only good news if trust and fit move within the following weeks.
- Watch the gap between endorsement and trust explicitly: When endorsement growth outruns trust growth quarter after quarter, as it has for Grok, Alexa+, and Google Assistant this year, the brand is spending equity it hasn't yet earned.
- Pair Demand Power with Future Power: Demand Power tells you what a brand commands today. Future Power estimates what its predisposition implies about tomorrow, converting a simple tracker into a forecast your finance team can actually use.
- Benchmark against the full competitive set, incumbents included
Right now, every legacy assistant is losing brand equity at the exact same time, and none of them have responded.
How Kantar tracks this
This is what Kantar BrandDynamics was built to do: an always-on read of Demand Power and the attribute architecture beneath it, validated against sales outcomes across thousands of brands, with True Signal delivering daily granularity at category scale. BrandZ then connects that read to the number a chief financial officer recognizes, by valuing the brand itself. Muse is a live case in point. Rather than waiting for a future study to catch up, it's already being added to the tracker as the category's newest entrant.
The AI companies that instrument their brand the way they instrument their models, continuously, quantitatively and against a competitive baseline, will see the turn coming. The rest will read about it in the demand numbers two quarters too late. The firms that treat brand equity as seriously as they treat model benchmarks are the ones building something that lasts past the next release cycle.
Take a self-guided demo of BrandDynamics today. For a custom read on where your brand sits against the full category, and what the attributes underneath it predict for the next two quarters, contact Jeff Herrmann, EVP Tech and Telco Industry Partner, North America.
