EPISODE · FRIDAY, MARCH 13, 2026

GPT-5.4 Cites Brand Sites 56% of the Time, Google Maps Gets AI, and the Responses API Changes Everything

GPT-5.4 now cites brand sites 56% of the time — up 7x from GPT-5.3. Pricing pages cited 34x more. Google launches Ask Maps with Gemini. OpenAI Responses API gives agents full computer access. Meta's Avocado model delayed to May. Anthropic study: AI impacts knowledge workers most.

Ready to Play

0:00 / 5:48

In This Episode

  • • GPT-5.4 cites brand sites 56% of the time — up from 8% in GPT-5.3 (7x increase)
  • • GPT-5.4 sends 8.5 fan-out queries per prompt vs just 1 for GPT-5.3
  • • Pricing pages cited 138 times in GPT-5.4 vs 4 in GPT-5.3 — a 34x increase
  • • Google launches Ask Maps: Gemini-powered chatbot for trip planning inside Google Maps
  • • OpenAI Responses API: agents get shell access, isolated containers, and reusable skills
  • • Meta's Avocado model delayed until at least May after failing internal benchmarks
  • • Anthropic study: AI impacts knowledge workers more than low-wage jobs
  • • Programmer and analyst hiring already slowing in AI-exposed roles

Show Notes

  • Introduction
  • GPT-5.4 cites brand sites 56% of the time — 7x increase from GPT-5.3
  • Google Maps Ask Maps — Gemini chatbot for trip planning
  • OpenAI Responses API — full computer environment for agents
  • Meta Avocado model delayed until May
  • Anthropic study: AI impacts knowledge workers most
  • Week in review — the compounding GEO advantage window
  • Closing

Transcript

[0:00] Introduction

Welcome to the AI Daily Digest for Friday, March 13th, 2026. Happy Friday — and we're closing the week with a story that should immediately change what you put on your website.

GPT-5.4 is now citing brand sites 56 percent of the time. That's up from just 8 percent in GPT-5.3. Google Maps just got its biggest redesign in a decade, powered by Gemini. OpenAI launched a Responses API that gives AI agents a full computer environment. And Meta's next flagship model has been delayed.

[0:35] The GPT-5.4 Citation Explosion

This is the most important GEO data point of the week, and possibly the year. A new study comparing GPT-5.4 and GPT-5.3 citation behavior found that brand sites now receive 56 percent of all GPT-5.4 citations — compared to just 8 percent in GPT-5.3. That's a 7x increase in one model generation.

The mechanism behind this is also significant: GPT-5.4 sends 8.5 fan-out queries per prompt, versus just 1 for GPT-5.3. The model is doing far more research before answering, and it's increasingly landing on brand-owned content as the authoritative source.

The most striking data point: pricing pages are now being cited 138 times versus just 4 in GPT-5.3 — a 34x increase. Commercial intent pages, which were previously ignored by AI models, are now among the most-cited content types.

[1:55] Google Maps Ask Maps

Google launched Ask Maps this week — a Gemini-powered chatbot built directly into Google Maps for trip planning and complex location questions. It's rolling out in the US and India on mobile first, with web support coming later.

If you have a local business or manage local SEO for clients, Google Business Profile optimization just became a GEO task, not just an SEO task.

[2:55] OpenAI Responses API

OpenAI launched the Responses API this week — a new API that gives AI agents access to a full computer environment, including shell commands, isolated containers, and automatic context management. It also includes reusable skills that package multi-step workflows so agents can be composed from modular building blocks.

[3:50] Meta Avocado Delayed

Meta's next flagship AI model, internally codenamed Avocado, has been delayed until at least May after falling short on internal benchmarks for reasoning, coding, and writing. Meta has already named its next model Watermelon.

[4:30] The Labor Market Signal

A new Anthropic study this week found that AI is impacting knowledge workers more than low-wage jobs — the opposite of what many economists predicted. Programmers, customer service representatives, and financial analysts are the most exposed roles. Hiring of younger workers in AI-exposed roles is already slowing.

[5:10] Week in Review

GPT-5.4 cites brand sites 7 times more than its predecessor. Google is embedding AI into Maps, Search, and every other surface. OpenAI is building the infrastructure for production agents. And GEO has moved from a niche tactic to a mainstream agency service line.

The brands that are investing in structured, entity-rich, AI-citable content right now are building a compounding advantage. The window to get ahead of this is still open — but it's closing.

Related Episodes

More episodes you might enjoy based on similar topics and categories.

AI Models & Research

Wednesday, March 18, 2026

GPT-5.4 Mini Costs Less Than a Stamp, Google Now Reads Your Gmail, and Why Faster Code Is Making Teams Slower

OpenAI releases GPT-5.4 mini ($0.75/M tokens) and nano ($0.20/M tokens) — production models designed for high-volume workflows. Google expands Personal Intelligence across Search, Gemini, and Chrome, pulling from Gmail and Photos to personalise results. AI overviews cut click-through rates by 58%, but AI-referred visitors convert at higher rates. Stripe's coding agents push 1,300 PRs/week using blueprint hybrid workflows. And the Atrophy Paradox: heavy AI use reduces cognitive engagement and makes output converge.

OpenAIGoogleSEOProductivityMarketing

7:10

AI Models & Research

Thursday, March 5, 2026

Gemini 3.1 Flash Lite Beats GPT-5 Mini, Claude Code Goes Voice, and Bing Makes GEO Official

Google launches Gemini 3.1 Flash Lite — 2.5x faster than previous version, outperforming GPT-5 mini and Claude 4.5 Haiku at $0.25/1M tokens. Claude hits $2.5B run-rate and overtakes ChatGPT in App Store. Apple's MacBook Neo launches at $599 with A18 Pro chip. Bing officially adds GEO to webmaster guidelines with opt-out mechanism. DeepSeek V4 drops with 1 trillion parameters and 1M context window.

GoogleAnthropicOpenAILLMsProductivity

8:10

AI Models & Research

Monday, March 9, 2026

Claude Code Beats Cursor, 91.3% Structured Content Win, and Anthropic Sues the Pentagon

Claude Code hits $2.5B ARR surpassing Cursor. New research shows structured pages achieve 91.3% AI Mode citation match rate vs 39.3% for unstructured. Anthropic files federal lawsuits against Pentagon over safety guardrails. Stargate Abilene expansion abandoned. The Eligibility Era reframes SEO strategy.

GEOSEOAnthropicClaudeCoding

6:45

AI Models & Research

Saturday, March 7, 2026

GPT-5.4 Goes Agentic, Codex Finds 800 Bugs, and the 90.9% GEO Benchmark

OpenAI releases GPT-5.4 with 1M token context and 75% desktop automation. Codex Security finds 800 critical bugs in OpenSSH and Chromium. Samsung AI glasses enter wearables race. Netflix acquires Ben Affleck's AI filmmaking startup. GenOptima sets 90.9% AI recommendation rate as the new GEO benchmark.

OpenAIAgentsSEOSecurityLLMs

5:46