Today’s focus: Google’s May 2026 core update remains the main SEO story, but the bigger strategic shift is happening around AI search: agentic retrieval, preferred sources, AI citations, and ChatGPT’s emerging ad ecosystem.
TL;DR
Google’s second core update of 2026 is still rolling out, so ranking volatility should not be treated as final signal yet. The most important GEO takeaway today is that AI search is moving beyond simple retrieval into multi-step, agentic systems that plan, retrieve, evaluate, and synthesize before producing answers.
For publishers and brands, the direction is clear: generic content is losing leverage. Original research, trusted sources, third-party authority, strong entity signals, and answer-ready content are becoming more important than publishing volume alone.
Must-cover items
1. Google’s May 2026 core update remains the main SEO volatility story
Google’s May 2026 core update is the second broad core update of the year. The rollout is expected to take around two weeks, which means early traffic and ranking swings should be treated carefully rather than interpreted as final outcomes.
The practical takeaway is simple: do not panic-edit, delete, or noindex content based on first-week movement. Wait until the rollout is complete, then compare performance across at least a full week of post-rollout data.
Why it matters: Core updates can reshuffle rankings across many verticals. SEO teams should avoid reactive decisions and focus on whether content is genuinely useful, differentiated, and satisfying for real users.
Source: Search Engine Roundtable – Google May 2026 Core Update Is Rolling Out
2. ChatGPT upgrades to GPT-5.5 Instant and Thinking
OpenAI updated GPT-5.5 Instant and GPT-5.5 Thinking in ChatGPT and the API, with improvements aimed at clearer, more natural responses and better in-chat writing and coding output. Older ChatGPT-only models are also being retired, including GPT-4.5 and OpenAI o3.
This matters for AI search monitoring because model changes can affect answer style, citation behavior, retrieval interpretation, and how brands or sources are summarized within ChatGPT experiences.
Why it matters: Any serious AIO/GEO tracking system should log model changes. If visibility, citations, or answer quality shift after a model update, the model change itself may be part of the explanation.
Source: Releasebot – ChatGPT Updates & Release Notes
3. AI search is moving beyond RAG into agentic retrieval
A new Search Engine Land analysis argues that major AI search systems are moving beyond classic retrieval-augmented generation. Instead of pulling a few documents and generating a single answer, these systems increasingly decompose a query, plan sub-queries, use tools, retrieve across multiple passes, evaluate evidence, and then synthesize the final response.
This is a major shift for GEO. If an AI system uses five to twenty hidden retrieval steps before producing an answer, simple prompt scraping and citation counting only show the final layer of a much deeper process.
Why it matters: Content must now survive a multi-stage evaluation loop. That means entity clarity, structured evidence, source authority, unique data, and topical completeness matter more than isolated keyword targeting.
Source: Search Engine Land – Beyond RAG: Why every AI search platform is now agentic
4. The May core update strengthens the trust-over-volume thesis
Writesonic’s analysis of the May 2026 core update frames the shift as less about individual pages and more about trusted sources. The argument: as Google uses AI Overviews and AI Mode to synthesize answers, it must decide which sources are credible enough to influence those answers.
The analysis is not neutral academic research, so treat it as a vendor perspective. Still, the direction aligns with what many SEOs are seeing: generic, derivative content is more vulnerable, while expert-led content, original research, strong brand signals, and external citations are gaining strategic importance.
Why it matters: The winning SEO/GEO strategy is no longer “publish more.” It is “become a source worth trusting.” That requires original data, expert authorship, brand authority, and reputation signals beyond your own website.
Source: Writesonic – Google May 2026 Core Update: POV on Search’s Trust Shift
5. Preferred Sources expands into AI Mode and AI Overviews
Google is bringing Preferred Sources into AI Mode and AI Overviews. Users can select publications or websites they trust, and when those sources appear in AI-generated answers, Google can highlight them more clearly.
Google is also adding more article carousels, social/forum perspectives, and “Highly Cited” labels to help users identify influential or primary reporting. This is especially relevant for publishers, media brands, and expert-led sites.
Why it matters: Audience loyalty may become a direct AI-search visibility asset. Brands and publishers should consider asking readers to add them as a Preferred Source, especially if they publish timely, original, or expert content.
Sources: Search Engine Land – Preferred Sources in AI Mode and AI Overviews; 9to5Google – AI Mode and AI Overviews will highlight Preferred Sources
On the radar
- ChatGPT ads are becoming a real paid media surface. Writesonic’s vendor study reported that ChatGPT ad penetration jumped sharply on May 26, reaching nearly half of U.S. responses in its sample. Treat the numbers as directional rather than definitive, since the company sells AI visibility tools. Source
- Kevin Indig’s latest Growth Intelligence Brief shows a divergence pattern. SaaS visibility declined while local and jobs-related aggregators gained visibility, reinforcing the idea that Google is routing more queries toward large review and aggregator platforms. Source
- Core-update volatility remains hard to attribute. Digital Applied notes that May’s core update overlaps with major Google I/O AI Search changes, making it difficult to separate ranking-system movement from AI Mode and interface changes. Source
- OpenAI is preparing conversion-focused ChatGPT ads. Search Engine Land reports that OpenAI is adding conversion tracking, a pixel, and Conversions API support for advertisers. Source
- CNN is suing Perplexity. Reuters reports that CNN alleges Perplexity unlawfully copied and distributed its content. This adds more pressure to the publisher-versus-AI-search legal conflict. Source
- FAQ content is still useful, but only when grounded in real demand. Search Engine Land recommends mining Google Search Console, People Also Ask, internal search, customer-service data, and Reddit to identify questions worth answering. Source
Original research spotlight
ChatGPT ad-penetration study – Writesonic
Key finding: Writesonic reported that ChatGPT ad penetration jumped from almost zero to more than 14% overall in its tracked sample, with U.S. responses reaching nearly 48%.
Methodology: The company says it analyzed 363,129 ChatGPT responses over 31 days and compared its findings with another public study that showed similar U.S. results.
Credibility note: The sample size is useful, and the cross-check with a second study improves confidence. But this is still vendor research from a company with a commercial interest in AI visibility monitoring, so treat the figures as strong directional evidence rather than final market truth.
Source: Writesonic – ChatGPT Ads Went From 0.03% to 14% Overnight
Growth Intelligence Brief #19 – Kevin Indig
Key finding: Kevin Indig’s latest brief reports a 7.8% decline in SaaS visibility over 30 days, while local search rose 44% and the jobs vertical gained 21.2%.
Methodology: The brief is based on Indig’s proprietary SEO Site Index, which tracks thousands of domains across multiple verticals, as well as AI prompt count data over a similar window.
Credibility note: This is proprietary data, so the full methodology is not independently auditable. Still, Indig’s dataset is useful for spotting directional movement across categories.
Source: Growth Memo – Growth Intelligence Brief #19
Quiet day?
No. It was not a quiet day. The core update is still active; Google’s AI Search interface is shifting; AI answer engines are becoming more agentic; and ChatGPT’s ad layer is moving from experiment to monetization infrastructure.
Editorial takeaway
The direction for SEO and GEO is clear: the search game is becoming less about isolated rankings and more about being selected as a trusted source inside AI-mediated discovery. The highest-leverage work now is original research, expert content, source authority, brand reputation, and clear entity-level positioning across the open web.