How AI Search Is Recommending AI Chatbots: Monthly Trends

Executive Summary

WATI remains the clear leader in AI chatbot recommendations, holding 22.1% valid recommendation coverage in September 2026, down from 26.2% in the July baseline. Yellow.ai follows at 6.9% coverage (down from 11.1% in July), with Interakt third at 6.0% (down from 9.3% in July). The gap between WATI and the next brand stands at 15.2 points in September, compared with 10.1 points in August.

The category recorded no significant movement on the primary coverage metric this month. All eight tracked brands are classified as stable, and the benchmark describes September as a quiet month overall. Within that quiet baseline, the most visible pattern is a narrowing of the field immediately behind the leader: Yellow.ai and Interakt both moved down from their August readings, compressing the spread between second, third, and fourth place even as the gap to the leader widened.

Two brands showed zero valid recommendation coverage this month. Engati moved from 0.4% coverage in both July and August to 0.0% in September, and Gallabox moved from 1.3% in both prior months to 0.0%. Haptik declined from 1.3% in July to 0.4% in August and then to 0.0% in September. Geta.ai remained at 0.0% across all three tracked months, with no recorded valid recommendations in the benchmark's tracked period.

Each monthly run begins with a collection of prompt-surface observations across the benchmark's defined AI/search surface universe. September 2026 began from 454 total prompts (305 unique questions). Of those, 445 mentioned a tracked brand or competitor; 340 were relevant and 105 were irrelevant. The public metrics use 217 qualified observations that survive both qualification stages. The July baseline began from 397 prompts (308 unique questions) and produced 225 qualified observations; August began from 491 prompts (385 unique questions) and produced 236 qualified observations.

Key Findings

Benchmark Context

The report separates the raw collection universe from the qualified analysis set. Brand-level recommendation percentages are calculated within the qualified benchmark set.

Benchmark-Level Metrics

The metrics below describe the qualified observation set as a whole, before breaking results out by brand.

August 2026 sat between the baseline and current months with 236 qualified observations, a recommendation-shaped answer share of 18.2%, and a valid recommendation shortlist share of 28.8%. The shortlist share moved downward across each of the three months, from 38.2% in July to 22.6% in September.

AI Recommendation Trend

The Leader's Position Holds as the Field Behind It Narrows

No brand's movement on the primary coverage metric exceeded the benchmark's normal month-to-month variation this month. The category-level change reflects the combination of several smaller movements, with the most concentrated pressure on the brands directly behind the leader. WATI's decline of 4.1 points from the July baseline was partly offset by a 1.8-point move up from August, leaving the leader's September position well above the rest of the field while four of the seven other tracked brands registered coverage at or near zero.

What Changed This Month

WATI: Leader by a Widening Margin

WATI's valid recommendation coverage moved from 26.2% in July to 22.1% in September, a net decline across the full window, though the current month sits above the August reading of 20.3%.

WATI's raw mention presence rose to 64.5% in September, up from 61.8% in the July baseline, meaning the brand appeared in nearly two-thirds of qualified observations. Its rank-one recommendation rate fell from 13.3% in July to 5.5% in September, a decline of 7.8 points. Its top-three rate moved from 17.8% to 12.4%.

The distinction to notice: WATI is more visible in September than it was in the July baseline, but it is being placed at the top of recommendations less often. The brand still converts presence into valid recommendations at 22.1%, and with the rest of the field declining, that conversion leaves it further ahead of second place.

Highest-priority diagnostic: Which prompts shifted WATI away from the rank-one position, and which brands are appearing in its place within the top three?

Yellow.ai and Interakt: The Middle Narrows Behind the Leader

Yellow.ai's valid recommendation coverage fell from 11.1% in July to 6.9% in September, a decline of 4.2 points, and has now declined for two consecutive months (11.1% to 10.2% to 6.9%). Interakt fell from 9.3% to 6.0% over the same window, down 3.3 points, including a 3.8-point move from August to September alone.

Both brands maintained their raw mention presence across the period. Yellow.ai's presence was 25.8% in September, nearly flat against its 25.3% July baseline, while Interakt's presence rose from 25.8% to 26.7%. The decline sits in recommendation conversion rather than in visibility: both brands appear in roughly a quarter of qualified observations but are recommended less often.

The distinction to notice: Yellow.ai's top-three rate fell from 7.1% in July to 2.3% in September, while Interakt's top-three rate moved from 6.2% to 3.7%. Both brands remain present in the conversation but are losing placement in the recommendation slots that matter most.

Highest-priority diagnostic: Which competitor is capturing the recommendation slots Yellow.ai and Interakt held in July, and in which prompt categories?

Gupshup: Gradual Decline Across Three Months

Gupshup's valid recommendation coverage fell from 4.9% in July to 2.3% in September, a decline of 2.6 points across the window, with August at 3.4% between the two. The brand has now declined for two consecutive months.

Gupshup's raw mention presence held relatively steady, moving from 10.7% in July to 9.7% in September. Its rank-one rate moved from 0.0% in July to 0.5% in September (1 rank-one mention out of 217 observations), while its top-three rate declined from 1.8% to 0.9%.

The distinction to notice: Gupshup operates at small counts where a handful of observations move percentages materially. Its 2.3% coverage represents 5 valid recommendations in September, down from 11 in July, so the decline should be read as a directional signal rather than an established trend.

Highest-priority diagnostic: Which prompts stopped returning Gupshup in a recommendation position, and is the decline concentrated on particular AI surfaces?

Engati, Gallabox, Haptik: At Zero Coverage This Month

Three brands recorded 0.0% valid recommendation coverage in September. Engati fell from 0.4% in both July and August to 0.0%, with 2 mentions in September and no valid recommendations. Gallabox moved from 1.3% in July and August to 0.0%, with 4 mentions and no valid recommendations. Haptik declined from 1.3% in July to 0.4% in August to 0.0%, with 4 mentions and no valid recommendations in September.

All three brands retain some raw presence, meaning AI systems still mention them, but none of those mentions converted into a recommendation this month. The distinction to notice: presence without recommendation is a distinct signal from absence. Engati, Gallabox, and Haptik are still surfacing in answers but not being placed in recommendation positions.

Highest-priority diagnostic: For each brand, are the remaining mentions contextual citations rather than recommendation placements, and what would be required to convert presence into recommendation?

Geta.ai: Zero Coverage and No Presence Across the Tracked Period

Geta.ai recorded 0.0% valid recommendation coverage in September 2026, unchanged from 0.0% in August and 0.0% in July. Unlike Engati, Gallabox, and Haptik, which retained some raw mention presence even after falling to zero coverage, Geta.ai had no recorded presence in the September qualified observation set, matching the same zero-presence reading in August.

WATI records 22.1% valid recommendation coverage in September, while Geta.ai stands at 0.0%. The more important issue is what that gap means for Geta.ai's recommendation position: every other tracked brand in this benchmark, including the three brands that fell to zero coverage this month, still appeared in at least some qualified observations, while Geta.ai has now shown no presence for two consecutive tracked months. This is not a case of visibility failing to convert into recommendation — it is a case of no visibility at all, against a leader capturing better than one in five qualified observations and a field of seven other brands that all registered at least some presence at some point in the tracked window.

Highest-priority diagnostic: Why does Geta.ai have zero presence in AI-generated answers across two consecutive tracked months, and what would be required to establish even a baseline presence in the prompts where competitors are currently being surfaced?

Buyer-Intent Interpretation

In September 2026, all 217 qualified observations fell into the Brand Recommendation cluster. No observations were recorded in the Pricing & Value or Multi-Brand Comparison clusters, meaning the public benchmark cannot yet answer questions about how AI systems position brands on price, value, or direct head-to-head comparison. The evidence captures which brand is recommended, not the reasoning or trade-offs behind that recommendation.

Brand Opportunity Summary

The table below distills the current per-brand signal down to the single highest-priority diagnostic question for each brand.

The benchmark identifies where attention is warranted; a company-level analysis is needed to explain why.

Evidence Behind the Benchmark

The aggregate metrics are built from prompt-level observations (query, surface, recommendation outcome, rank, sentiment, and citations where exposed). Company-level analysis can go deeper into prompt, competitor, surface, and evidence patterns. Source presence is not automatically treated as proof of causation.

About This Benchmark

This report is part of the LLM Authority Index AI Market Discovery research program.

• AI Industry Market Discovery
• AI Industry Market Discovery Methodology
• AI Industry Market Discovery Metrics
• AI Industry Market Discovery Standards

Report-Specific Interpretation Notes

• Small-count movements: brands such as Engati, Gallabox, Gupshup, and Haptik operate at counts where a single observation changes percentages meaningfully. Read these as directional signals, not established trends.
• Qualified denominator: all brand-level percentages are calculated against the qualified observation count (217 in September), not the raw collection size (454 prompts). This separates the analysis set from the broader collection universe.
• Month-over-month movement identifies changes worth investigating. It does not by itself establish the cause of those changes, and this month's category-level pattern was quiet, with no brand exceeding the benchmark's normal month-to-month variation on the primary coverage metric.

Next Step

The Public Benchmark Shows Where a Brand Is Winning or Losing. A Company-Level Audit Shows Why.

Beneath the aggregate percentages sit the questions that matter: which high-intent prompts are won, which competitor takes the recommendation when a brand loses, what attributes AI systems associate with each option, and which external sources shape those answers. The public benchmark identifies the surface-level movements; it does not expose the underlying drivers.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy.

Request an AI visibility audit