In November 2025, AMA New Orleans hosted AI is Changing SEO: Challenges, Opportunities, and Strategies. Moderated by AMA NOLA board member Petera Diaban, the panel included three of the top SEO experts in Louisiana: Brian Reed of FSC Interactive, Paula French of Search Influence, and Lauren Walter of Online Optimism.

The panel tackled everything from how webmasters can improve their visibility in LLMs (and how to measure their performance) to the role social media plays, and will play, in how brands are mentioned in LLMs and AI answer boxes.

The Big Question: Is AI really killing (traditional) search traffic?

The short answer is no. As panelist Lauren Walter cited, a study by SE Ranking found AI platforms account for just 0.15% of global web traffic, compared to nearly 50% for organic search. Ahrefs found similar results, reporting AI drives only 0.1% of total referral traffic.

The long answer is no, but it’s having an effect.

Brian commented that search marketers have panicked about a lot of things that were supposed to “kill” SEO but didn’t (like voice search), but the difference is AI is embedded in SERPs and sometimes even appears before organic results, so there are some cases of AI impacting CTRs. (But likely not as much as people think.)

Paula agreed there’s been a bit of an impact, but mainly on more established SEO clients. That said, there’s still much more happening on Google than ChatGPT or other AI platforms.

What are Google’s motivations with search and AI?

The next panic is over Google’s long-term goals for search and AI. We asked the panel if Google’s intention is to kill organic CTRs by answering search queries via AI overviews and answer boxes and keeping them on Google SERPs indefinitely.

The panelists disagreed.

Brian argued since Google makes the bulk of its revenue from ads, it’s not in Google’s best interest to tank CTRs. “We’d have way more AI overviews now if they could figure out how to better monetize them.”

The Challenge: Adapting to a changing landscape

Ranking in Google SERPs is different from being mentioned in AI prompts. The panel addressed what changes matter most to marketers.

AI crawlers vs. search engine crawlers

Lauren put it best when explaining page-level vs. passage-level relevance. “Google is trying to decide if it wants to rank your entire page, whereas LLMs are trying to decide to pull a specific passage.” For both search engines and LLMs, she recommends using schema markup to help bots understand the content on your site, using clear and structured headings, and breaking up content into easily digestible chunks.

Brian was optimistic LLMs would have a positive impact on how site content is structured. Search engines have historically preferred more long-winded, comprehensive, topically-based content with lots of keywords (and lots of potential keyword rankings), whereas LLMs look at content at a more contextual level. He used an example of searching for recipes. While it may be SEO best practice to include way more content than just the recipe (e.g. a history of the ingredients, where did the recipe originate, etc.), LLMs won’t care about superfluous content.

Importance of original content

The panel was asked if AI can generate content on any topic in seconds, should marketers even bother using resources to create original content themselves?

“Original content is more important than ever,” said Paula. If you want to be recognized as an authority by Google or LLMs, you have to give your own unique perspective. Novel content is more likely to be trusted and sourced by search engines and LLMs; you need to stand out from the slop everyone else is creating with AI.

Lauren agreed. “If you’re just regurgitating what’s already on page one of Google or already in AI overviews, why would your content appear in place of what’s already there?” She recommended incorporating originality. “Original data can be very difficult to compile, and everyone wants to be able to cite relevant statistics related to their field.”

Combatting misinformation from LLMs

AI can get things wrong, whether from “hallucinations” or from misinformation on the internet that gets sourced by LLMs.

Brian didn’t see hallucinations as a big issue, but warned of malicious activity that could affect how LLMs discuss your brand. “This opens the opportunity for people to spam the internet with bad things about you that will show up in LLMs.” He suggested monitoring your brand in LLMs regularly: asking them what they know about you, how you’re talked about in public spaces, your reviews, etc.

Paula mentioned that scouring the web for incorrect or inconsistent business information was important even before ChatGPT existed. You can try asking the LLM where it’s getting the (incorrect) information, but it’s better to go straight to Google to find the source of the misinformation.

Lauren added that in ChatGPT you can downvote responses and provide context for what you didn’t like about a response or even offer a correction, but couldn’t vouch for its effectiveness.

The Big Picture: Thinking beyond websites

With news that Facebook and Instagram posts are being indexed in search (and TikTok posts long before that), we asked if search marketers should focus more attention on building offsite content (on social media) rather than onsite content generation.

“Social media SEO” vs. your website’s SEO?

Paula saw value in offsite content generation on social media. “We’ve focused more on content repurposing since big sites tend to get referenced in both organic search and LLMs. We frequently see YouTube and Reddit, in addition to the social media sites mentioned, as citation sources.” Going back to the question of losing traffic, Paula mentioned seeing less traffic on blog posts (“informational searches”), but repurposing blog posts on larger sites, which users are starting to use more similarly to Google, is a great way to combat the loss in site traffic.

Lauren recommended focusing on where your audience is. “It’s not really an either-or, but social media sites are great for augmenting SEO.” Google is still a larger share of referral traffic than Facebook or Instagram, but if your audience skews younger, then it’s possible you’d want to focus more on TikTok than Google, even if that’s not reflected in larger global trends.

Measuring What Matters: How to track AI’s impact on brand visibility

Is there any way to accurately measure brand visibility in AI? And are there any tools that you trust?

Preferred tools

Brian mentioned Semrush (NOTE: this was before news that Semrush was acquired by Adobe), but suggested going straight to the log files to see how often AI bots are visiting your site, and to which pages.

Paula said that while no tool is perfect, her company Search Influence analyzed eight of the best AI tracking tools and found that Scrunch AI and Profound were their favorites. She also advised against signing any long-term contracts with any particular tool.

Lauren seconded Semrush and Profound and offered Otterly AI as a more affordable option.

Improving brand authority in AI overviews

“One thing we’re seeing is these AI platforms heavily favor community forums and discussions. Reddit especially, but also Quora,” said Lauren.

Brian agreed that LLMs are trying to get public sentiment for various search queries. “You want to have reviews in places where people care about reviews,” he added. Whether it’s TripAdvisor, Yelp, Trustpilot… wherever your product or service is reviewed, LLMs are going to look there and make decisions based on those reviews.

Looking Forward: Getting ahead of the next disruption

Finally, we asked the panelists for their AI predictions.

And that’s where this recap ends. You’ll have to watch the webinar (embedded above or watch on YouTube here) to get their predictions!