SEO 101 Ep 525: Google AI Mode Threatens Web Traffic and Publishers

Release Date: March 18, 2026
Hosts: Ross Dunn (CEO, StepForth Web Marketing), Scott Van Achte (Senior SEO, StepForth)
Podcast: SEO 101 - WMR.FM

Episode Overview

This episode of SEO 101 dives into the pressing concerns surrounding AI advancements in search, particularly how Google’s AI Mode is impacting web traffic, publishers, and the SEO landscape as a whole. The hosts break down the threat posed by answer engines and AI overviews, discuss the evolution of personalized search, reflect on content trends and penalties (like listicles), and highlight alarming issues around misinformation propagating through AI-driven systems.

Key Discussion Points and Insights

1. Early Technical Banter & Industry Humor

• Lighthearted conversation on “Blueski” vs. Bluesky and the pain of repetitive hold music in customer support systems.
• The hosts muse over whether annoying hold music is a conspiracy:
  • Scott (03:11): "I think they do it on purpose to tick people off so they just get angry and hang up."

2. OpenAI’s Ads Expansion in ChatGPT (04:20)

• News: OpenAI is testing an Ads Manager for ChatGPT, offering marketers real-time control. Previously, advertisers depended on manual changes and weekly CSV data.
• Critical Take:
  • Scott (05:23): "It’s insane that...they even launched it in beta without that capability [self-serve updates]."
  • Ross (06:01): Explains it’s part of the “iterate as you go” product trend: "You just run with it first and then iterate as you go and improve. Improve or abandon it."

3. Google AI Mode – A Threat to Publishers (06:12)

• Jim Lanzone (Yahoo CEO) recently labeled Google’s AI Mode the “biggest threat to web traffic,” warning that answer engines are starving publishers of clicks.
• Stats Highlighted (07:15-08:39):
  • Referral traffic from traditional search engines declined by 60% for small publishers, 47% for medium, and 22% for large over two years (Chartbeat data).
  • Page views from Google Search and Discover fell 34% and 15% between Dec 2024–Dec 2025.
  • AI-driven chatbots like ChatGPT increased referrals by 200% but still represent under 1% of publisher pageviews.
• Ross: “That’s a huge drop. Holy smokes... To financially survive on it, it gives me the willies. I feel for them.” (08:39)

4. Google’s Personal Intelligence Expansion (09:19)

• Google rolls out “personal intelligence” in its AI-Mode results, Gemini app, and Gemini in Chrome Beta.

• Implications:

  • Personalization uses personal data (Gmail, Photos) for tailored responses (e.g., shopping recommendations, receipts for troubleshooting, travel history).
  • This makes SEO harder – results heavily customized, harder to track rankings.

• Privacy Details:

  • Opt-in only; models don’t train “directly” on Gmail or photos.
  • Prompts/responses may be used to improve systems (like “this call may be monitored...” warnings).
  • Personal Intelligence features are currently ad-free but ads expected in the future.

• Memorable Quote:

  • Scott (13:35): "Yeah, it would be really creepy if...in the results is an email from your Gmail account or a photo from your personal, you know, Google Photos... I don't want to see my stuff show up in search results unless it’s public."
  • Ross (14:03): Reminisces about Google Desktop Search and its privacy quirks.

5. The Demise of Google’s “What People Suggest” in Health Search (15:27)

• Google removes the “What People Suggest” health feature, which used AI to organize health discussions from forums/social (Reddit, X, Facebook) for reliable information.
• Scott quips (15:27): “Reliable and medical doesn’t work this month.”
• Google claims removal is not due to safety or quality.

6. Google Penalizing Self-Promotional Listicles (16:22)

• Building on prior coverage, Scott summarizes Lily Ray’s recent findings:

  • Google appears to target sites whose “best of” listicles are thinly veiled self-promotion (often with the agency/site itself at #1).
  • January 2026 updates sharply decreased rankings and AI search citations for such sites.
  • Mass-produced, self-promoting listicles may be suppressed, while smaller/less obvious offenders persist.

• Key Advice:

  • Focus on honest, expert, and real-world-tested listicles, not mass-produced or always self-promoting ones.

• Memorable Moment:

  • Ross, laughing (18:09): “One of the more established news sites in our industry always makes me laugh when they do the top SEO podcasts because sure enough, number one’s always theirs.”

7. SEO Tests Show Spread of Misinformation is Trivial (18:34)

• Experiment: SEO John Goody creates a fake story about a “March 2026 Google Core Update”—it never happened.
• Despite big outlets ignoring it, small SEOs pick up the story. Google’s search and AI overviews start to rank and repeat this fake information, leading others to cite it, “proving” the hallucination via repetition.
• Example: Techbytes fabricates more details, referencing made-up filters and algorithm names.
• Takeaways:
  • AI workflows need validation.
  • Most readers don’t fact-check; AI overviews amplify misinformation.
  • New EU law (2025) requires engines to fact-check results, but Google won’t comply.
• Scott (20:23): “Maybe it’s time that they do [fact-check]. It’s going to get so much worse. This is nothing yet.”
• Ross (21:42): Best step: “Act on flagged sites that are regularly posting misinformation... and ban them... I would praise Google up and down if they did that. Man, that would be a huge... transform my thought of Google. So probably won’t happen, but...”

Notable Quotes

• Scott (03:11): "I think they do it on purpose to tick people off so they just get angry and hang up."
• Ross (08:39): "That’s a huge drop. Holy smokes... To financially survive on it, it gives me the willies."
• Scott (13:35): "Yeah, it would be really creepy if...in the results is an email from your Gmail account or a photo from your personal... I don’t want to see my stuff show up in search results unless it’s public."
• Scott quoting Lily Ray (17:00): “Plenty of sites are still getting away with this... But I wonder if the January 21st update wasn’t Google testing a way to demote this for a small batch of the worst offenders."
• Ross (21:42): "Act on flagged sites that are regularly posting misinformation... and ban them... I would praise Google up and down if they did that."

Timestamps for Key Segments

• 01:35 – Episode start, casual banter on aging/notes
• 03:15 – Commentary on customer service/hold music
• 04:20 – OpenAI’s ChatGPT Ads Manager
• 06:12 – Yahoo CEO: Google AI Mode as threat to web traffic
• 09:19 – Personal Intelligence in Google AI Search
• 15:27 – Removal of “What People Suggest” in health
• 16:22 – Google penalizing self-promotional listicles
• 18:34 – Experiment: Misinformation easily ranks via AI
• 22:01 – “Ban misinformation sites” debate

Summary Takeaways

• AI-powered answer engines (like Google’s AI Mode and ChatGPT) pose a significant threat to publisher traffic, impacting especially small publishers.
• Personalization in search will make SEO work more complex, as individualized results are harder to monitor and audit. Privacy remains a major concern.
• Listicles abusing self-promotion are increasingly targeted by Google’s updates. Genuine, honest content is more sustainable.
• Misinformation spreads rapidly through AI-driven search and AI overviews; without rigorous validation, fake news can propagate and become “true” in search.
• The future of SEO and publishing requires agility, authenticity, and caution in the age of rapidly iterating AI systems.

For further discussion, join the SEO101podcast Facebook group.