The GEO Show copertina

The GEO Show

The GEO Show

Di: Paris Childress - GEOforge
Ascolta gratuitamente

Offerte di stagione | 0,99 €/mese per i primi 3 mesi

A seguire 9,99 €/mese – si applicano condizioni. Puoi disdire mensilmente.
All things Generative Engine Optimization (GEO). A breakdown of all that's happening in the world of AI search.© Hop AI 2026 Economia Marketing Marketing e vendite
  • AI Engines Disagree on the Top Bank 43% of the Time
    Oct 21 2026

    AI engines disagree on which bank to recommend first 43% of the time. Paris Childress covers this week's GEO news, with one theme: AI visibility monitoring is becoming a commodity, and the real fight is moving to execution, methodology transparency and oversight of AI agents.

    In this episode:

    • Profound and Bazaarvoice partner to connect AI visibility with customer reviews, with research on about 40 million ChatGPT shopping product mentions.
    • geoSurge launches a free desktop tracker for ChatGPT, Claude and Gemini. Paris calls monitoring table stakes.
    • 43% disagreement. PRecious Communications analyzed 2,740 AI responses on Southeast Asian banking across 6 countries and 8 languages.
    • AccuRanker recalculates share of voice, cutting the median tracked domain 48% with no change in rankings or traffic.
    • Different raises a $48M seed round, and Salesforce agrees to acquire Braze for about $5.66B.
    • Tenable opens its security intelligence to AI agents through MCP, and Paris explains why software is going headless.
    • OpenAI discloses two incidents where agents weakened monitoring safeguards.
    • Ahrefs Brand Radar gets a TechRadar review, and Paris explains why GEOforge runs every prompt 30 times.
    • SPM Communications puts GEO at the core of its PR offering.

    Watch on YouTube: https://www.youtube.com/watch?v=tHWoLuVF8qA

    The GEO Show is hosted by Paris Childress, founder of Hop AI and co-founder of GEOforge. getgeoforge.com

    Mostra di più Mostra meno
    16 min
  • 84% of AI Citations Come From Earned Media, Not Your Site
    Oct 20 2026

    84% of AI-cited sources come from earned media, according to Muck Rack research highlighted in Force4's 2026 Market Influence Report. Paris Childress covers this week's GEO news, with one theme: the GEO market is splitting into measurement, activation and agent trust.

    In this episode:

    • Getfluence and Meteoria partner to connect AI visibility measurement with earned media placement.
    • 84% of AI citations are earned media. Force4's report cites Muck Rack data, and Paris argues PR outreach may be the most important part of GEO.
    • Agent commerce is ahead of deployment. Massive found published UCP profiles on only 11 of 36 storefronts tested.
    • Proof and Circle plan identity-gated agent payments, so people can trust their agents to buy.
    • Microsoft adds risk-adaptive security controls for enterprise AI agents.
    • Rank Control joins the measurement-to-execution trend in GEO tools.
    • Tracker methodology matters. One run per prompt per day is basically a coin flip.
    • GEO market size estimates for 2026 range from $160.9M to $4.91B, a 30.5x spread.
    • HubSpot brings AI search education to mainstream marketing teams.

    Watch on YouTube: https://www.youtube.com/watch?v=dQxpNYJH9pY

    The GEO Show is hosted by Paris Childress, founder of Hop AI and co-founder of GEOforge. getgeoforge.com

    Mostra di più Mostra meno
    14 min
  • The ChatGPT Your Tools Track Isn't the Real One
    Oct 19 2026

    Welcome to Episode 65 of The GEO Show, the daily podcast on Generative Engine Optimization: how brands earn visibility, mentions and citations inside AI answers.

    Paris Childress, founder of Hop AI and co-founder of GEOforge, covers this week's GEO news. The big theme: AI visibility measurement is drifting away from what users actually see, while AI agents move from recommending vendors to negotiating and transacting with them.

    In this episode:

    • 📊 The ChatGPT your tools track isn't the real one
      Researchers tested 80 questions in English and Chinese. For GPT-5.5 Instant, Google Search results plus OpenAI's Responses API recovered only 26.2% of the exact pages and 48.7% of the source sites the ChatGPT app cited. The study did not test commercial trackers, but most visibility tools measure through the API.
    • 🤖 Parallel agents in GPT-6 Pro
      OpenAI now lets several agents work on parts of one task at once, which prompt-based measurement may not capture.
    • 🌐 A six-engine citation index
      The Machine Relations Index counts 31,721 cited domains across 27,227 answer runs and 200,158 source events from May 10 to October 18, led by Reddit, YouTube, LinkedIn and Forbes.
    • 🤝 Anthropic and AI agent negotiations
      The Financial Times reports that Anthropic is consulting legal specialists as it explores how AI agents might negotiate terms. This is reported, exploratory work, not a launched service.
    • 🧰 The GEO tool market keeps growing
      AiSightKit lists 134 tools and platforms, and G2's AI search software category has more than 650 listings.
    • 🏨 AI visibility standards for hotels
      The AI Hospitality Alliance released 15 AI visibility indicators for hotels, including booking capability.
    • 🔍 Cloudflare, OpenAI and agent transactions
      A Cloudflare and OpenAI pilot on about 1% of Cloudflare's network surfaced hundreds of millions of previously unknown pages, though it does not show that those pages were cited. GoDaddy, Infoblox, Google and Cloudflare also demonstrated a prototype for trusted agent transactions.

    Subscribe to The GEO Show on Apple Podcasts, Spotify or YouTube for a new episode every day.

    💬 Question: Does your AI visibility tool measure the ChatGPT app or the API?

    Watch on YouTube: https://youtu.be/YHz3Z58vTjU

    Mostra di più Mostra meno
    10 min
adbl_web_anon_alc_button_suppression_t1
Ancora nessuna recensione