Pichai deflects Gemini 3.5 Pro delay with Gemini 4 tease

Key Takeaways

• Google CEO Sundar Pichai avoided discussing the Gemini 3.5 Pro delay during Q2 earnings call, pivoting to Gemini 4 plans instead
• Bloomberg reports Gemini 3.5 Pro is months late due to coding performance falling short against OpenAI and Anthropic
• Analysts warn monthly model releases could burden CIOs with testing, governance, and version management costs

Sundar Pichai dodged questions about Google's delayed Gemini 3.5 Pro during the company's Q2 2026 earnings call on July 23, steering the conversation toward Gemini 4 and a promised monthly release cadence instead. The deflection came one day after Google unveiled Gemini 3.6 Flash and 3.5 Flash Cyber but offered no update on its flagship reasoning model, which was originally promised for June.

What happened during the earnings call?

When Barclays analyst Ross Sandler pressed Pichai on how Google planned to compete in the frontier AI race, the CEO pivoted. "We are creating a baseline on top of which you will see us rapidly iterate on subsequent model releases. And so picking up pace and releasing models almost at a monthly cadence is part of our road map as we are building Gemini 4 as well," Pichai said.

JPMorgan Chase analyst Douglas Anmuth had asked a similar question moments earlier: whether Google was releasing frontier models frequently enough to keep pace with OpenAI and Anthropic. Pichai responded that Google remained confident at the frontier and was investing heavily in a larger Gemini 4 base model. He did not address Gemini 3.5 Pro directly.

Why is Gemini 3.5 Pro delayed?

Google introduced the Gemini 3.5 family at its annual I/O conference with a June release target for the Pro variant. That timeline has slipped by more than a month. Bloomberg reported that the model's coding performance is falling short of internal expectations, particularly when benchmarked against competing models from OpenAI and Anthropic.

The gap matters. Coding tasks have become a primary use case for enterprise AI deployments, from code generation and review to debugging legacy systems. A reasoning model that underperforms on code is a hard sell to engineering teams already comfortable with alternatives.

What does this mean for CIOs evaluating AI platforms?

The delay has not triggered an exodus of existing Google Cloud customers, according to Bhupendra Chopra, chief revenue officer at IT consulting firm Kanerika. High switching costs and multi-model architectures have kept enterprises in place. But it has made CIOs more cautious about new commitments.

Pichai's promise of monthly releases introduces a different problem. Sanchit Vir Gogia, chief analyst at Greyhound Research, warned that faster cadences will require CIOs to invest more heavily in testing, governance, and version management to safely adopt updates.

Pareekh Jain, principal analyst at Pareekh Consulting, put it bluntly: enterprises will only embrace rapid releases if each version delivers measurable improvements in performance, cost, or safety. "The challenge for CIOs is not just keeping up with model releases; it's deciding whether each new version is worth the cost of validating it," Jain said.

How does Google's AI roadmap compare to rivals?

OpenAI and Anthropic have both shipped reasoning-focused updates in recent months that specifically target developer workflows. Google's delay puts it in catch-up mode precisely when enterprises are locking in AI platform decisions for 2027 planning cycles.

Is monthly model release sustainable for enterprises?

A monthly cadence sounds aggressive. Most enterprises cannot retrain workflows, update prompts, and regression-test integrations every 30 days. The cost of validation alone could offset any performance gains from incremental model improvements.

The more realistic pattern is that enterprises will adopt a subset of releases, likely one per quarter, treating intermediate updates as optional. That shifts the burden to Google: each release must be backward-compatible enough that skipping versions does not break production systems.

Frequently Asked Questions

When will Google release Gemini 3.5 Pro?

Google has not provided a new release date. The model was originally promised for June 2026 but remains delayed as of late July 2026.

Why is Gemini 3.5 Pro delayed?

Bloomberg reports the model's coding performance falls short of internal expectations, particularly compared to competing models from OpenAI and Anthropic.

What is Gemini 4?

Gemini 4 is Google's next frontier AI model currently in development. Pichai mentioned it during the July 23 earnings call but provided no release timeline.

Will Google really release AI models monthly?

Pichai said releasing models at an almost monthly cadence is part of Google's roadmap, though analysts question whether enterprises can validate updates that frequently.

Should enterprises wait for Gemini 3.5 Pro or use alternatives?

Most analysts recommend multi-model architectures that allow enterprises to use OpenAI, Anthropic, or other providers while waiting for Google's delayed flagship.

Source: Computerworld