Gemini AI Agents Resolve 75% of Google Ads Support Queries

What happened

Alphabet used its Q2 earnings call to highlight a significant operational milestone: AI agents powered by its Gemini models are now autonomously resolving 75% of Google Ads support queries, without human intervention. The disclosure marks one of the most concrete, publicly cited benchmarks for large-scale AI-driven customer service automation from a major technology company.

Beyond the support desk, Alphabet outlined an expansive strategy to embed Gemini across the entire customer lifecycle. That includes conversational search, agentic shopping and bookings, advertising optimisation, and sales enablement — signalling that the company views AI not as a bolt-on feature but as the connective tissue of its customer-facing operations.

The announcement comes as competition in enterprise AI intensifies, with Alphabet positioning Gemini as its answer to rivals deploying large language models in customer engagement and service automation. The Q2 earnings context gives the 75% resolution figure particular weight: it is a performance claim tied to a financial reporting event, lending it a degree of accountability that typical product announcements rarely carry.

Why it matters

A 75% autonomous resolution rate at Google Ads' scale represents a meaningful inflection point for how the industry thinks about AI in service design. For CX leaders, the figure reframes the conversation: the question is no longer whether AI agents can handle complex, intent-driven support interactions, but how quickly organisations need to redesign their service architectures around that capability. The behavioral economics angle is equally significant — customers who receive fast, accurate resolutions without being transferred or asked to repeat themselves experience a dramatic reduction in effort, the single strongest driver of loyalty in high-frequency service contexts.

Alphabet's broader framing — AI across the full customer lifecycle — also challenges service designers to think beyond the contact centre. When the same underlying model handles search discovery, purchase assistance, post-sale support, and advertising performance queries, the boundaries between marketing, sales, and service dissolve. That convergence demands new organisational models and new ways of measuring customer experience end-to-end.

By the numbers

• 75% of Google Ads support queries are now resolved autonomously by Gemini-powered AI agents, as disclosed during Alphabet's Q2 earnings call.
• Q2 2025 earnings call was the occasion on which Alphabet made the milestone public, anchoring the claim to a formal financial reporting event.

The Renascence take

Most commentary on this news will focus on the headline resolution rate as proof that AI has "arrived" in customer service. That reading is too comfortable — and misses the harder strategic question the number actually raises.

A 75% autonomous resolution rate sounds like a victory, but it immediately makes the 25% that still reaches a human more consequential, not less. Those residual interactions are, by definition, the ones AI cannot yet handle — which means they are disproportionately complex, emotionally charged, or commercially sensitive. Customer-obsessed operators should resist the temptation to thin out human capacity in lockstep with automation gains; instead, they should be investing the cost savings from the 75% into dramatically upgrading the skill, empathy, and decision-making authority of the agents who handle the 25%. The behavioral principle here is contrast effect: when a customer finally reaches a human after an AI journey, their expectations and emotional state are heightened. That moment is a brand-defining interaction, not a fallback.