Strategy Marketing May 22, 2026 Our News RoomGoogle just released its second core update of 2026. And the most common question landing in my inbox right now is the same one I want to answer today — honestly, directly, and in plain language. “Hema, I use AI to help me write. Does that mean I am going to get penalised?” The answer might surprise you.
Google released its May 2026 core update on May 21, 2026 — the second core update of the year, following March 2026. The rollout takes up to two weeks to complete and affects search rankings across every sector.
But here is what Google has been consistently saying across every core update — and this one is no different:
“There’s nothing new or special that creators need to do for this update as long as they’ve been making satisfying content meant for people.”
That is Google’s guidance. Not “stop using AI.” Not “only publish human-written content.” Not “penalise everyone who used a tool to help them.”
That is the whole update. And it has enormous implications for how you think about AI-assisted writing.
No.
And here is the distinction that changes everything.
Using AI to help you write is not the problem.
Letting AI replace your thinking is.
These are two completely different things. And most of the panic around AI content is conflating them into the same conversation — which is both inaccurate and unhelpful.
Imagine you are baking a cake for your best friend’s birthday.
You use a stand mixer to help you blend the ingredients faster.
The mixer did not choose the recipe.
The mixer did not decide it was your friend’s favourite flavour.
The mixer did not add the extra thought, care, and intention you put into making something special for someone you love.
You did all of that.
The mixer just helped you do it faster — and more consistently — than you could have done by hand.
That is AI done right. The mixer. Not the baker.
Now imagine a different scenario. You tell the mixer what flavor you want, the mixer produces a generic cake from a standard template, and you hand it over without tasting it, without adjusting it, without adding anything of yourself to it.
That cake might look fine. It might even taste acceptable. But it was not made for your friend. It was made for anyone. And people — including Google — can tell the difference.
Here is the pattern that Google’s core updates are specifically designed to surface and deprioritise:
You open ChatGPT. You type a topic. You copy what comes out. You publish it without reading it properly, without adding your real experience, without checking whether it actually answers the question your reader is asking. You do this at scale across dozens of pages.
The result is content that is technically coherent but genuinely hollow. It covers the topic without illuminating it. It answers the question without actually helping anyone. It sounds authoritative without being earned.
Google’s algorithms — increasingly sophisticated at identifying genuine helpfulness versus surface-level adequacy — flag this content as unsatisfying. AI platforms like ChatGPT and Perplexity, which cross-reference Google’s trust signals when building their recommendation patterns, similarly deprioritise it.
The content exists. It ranks nowhere useful. It helps no one. And it damages your brand in ways that take months to recover from.
Here is the pattern that Google rewards — and that AI platforms cite:
You bring your expertise. Your real client stories. Your genuine perspective on a problem you have seen dozens of times across your sector. The specific nuance that only someone who has done this work for years actually understands.
You use AI to help you structure your thinking. To sharpen a draft. To fill gaps in coverage. To check whether your explanation is clear to someone who does not already know what you know.
You review every word. You add your voice. You remove what does not sound like you. You make sure the person reading it actually gets what they came for.
You publish something that only you could have written — just faster, better structured, and more complete than if you had started from a blank page alone.
That is AI as the mixer. You are still the baker.
Google loves this version. AI platforms cite this version. Real humans read this version — and call you.
Here is a data point from May 2026 that should inform every content decision you make this year:
Nearly half of all online articles are now AI-generated.
Half.
Which means the internet is flooding with content that sounds the same. Generic. Interchangeable. Written for no one in particular about everything in general.
Google’s May 2026 core update is specifically designed to surface the content that rises above that noise. The content that is genuinely satisfying. That was clearly written for a specific human being with a specific problem.
In that context — your real expertise, your authentic voice, your genuine perspective — has never been more valuable.
AI gives everyone the same starting ingredients. Your human experience is the differentiator.
Here is what most businesses are not yet tracking — and should be.
Google’s core updates do not just affect your Google rankings. They send ripple effects across every AI platform your buyers are using.
ChatGPT, Perplexity, Claude, and Google’s own AI Overviews cross-reference Google’s trust signals when building their recommendation patterns. When Google decides your content is genuinely satisfying — when it rewards your pages with higher visibility — that signal is absorbed into the broader AI ecosystem over time.
Conversely, when Google determines your content is hollow, generic, or designed for algorithms rather than people — that signal ripples outward too.
This is why Agentic Engine Optimization is not a separate discipline from content quality. They are the same discipline, viewed from two different angles.
The content that wins Google’s May 2026 core update is the same content that gets cited by AI platforms when your buyer asks the question you should be answering.
People-first. Genuinely helpful. Backed by real expertise. Validated by independent third parties who had no reason to vouch for you other than that you actually delivered.
That is the content that wins on both surfaces simultaneously.
In my book The AI Translator, I outline the 4 Vs framework — the four signals that both Google and AI platforms use to evaluate whether a brand deserves to be recommended.
Visible — Is your content structured so that both Google and AI platforms can find, read, and extract it efficiently? This means proper schema markup, clean site architecture, and consistent entity signals across every platform where you exist.
Validation — Do other humans independently trust and validate you? Reviews, citations, third-party press mentions, and case studies from people who have no financial reason to vouch for you. This is the signal that is hardest to fake — and the one both Google and AI platforms weigh most heavily.
Veracity — Can your claims be independently verified? Google’s E-E-A-T framework and AI platform recommendation systems both cross-reference assertions against available evidence. If the proof is not there, the recommendation does not happen.
Values — Do you show up consistently, treat people well, and produce content that is genuinely people-first over time? Consistency is a trust signal. Both Google and AI platforms reward it. Both penalise its absence.
Notice that none of the 4 Vs asks “did a human write every word of this content?”
They ask: is it genuine, validated, verifiable, and consistent?
AI-assisted content written by a genuine expert, grounded in real experience, validated by real client outcomes, and structured for real human readers — passes all four.
AI-generated content written for no one in particular, covering everything superficially, backed by no independent proof — fails all four.
The tool is not the issue. The intention and the expertise are.
If you want to use AI to help you create content that Google rewards, AI platforms cite, and real humans find genuinely useful — here are the principles to work from:
Start with your expertise, not with AI. Before you open ChatGPT, write down the three things you know about this topic that most people get wrong. The specific nuance. The real-world complication. The thing you have learned from clients that no generic article ever captures. That is your content. AI helps you structure and express it — not generate it from scratch.
Use AI for structure and clarity, not for substance. AI is exceptional at helping you organise complex information, identify gaps in your explanation, and check whether your language is accessible to your target reader. Use it for those things. The substance — the actual knowledge, the real experience, the genuine perspective — must come from you.
Read every word before you publish. If you would not be comfortable saying every sentence out loud to a client in a meeting, it does not belong on your website. AI sometimes produces technically correct but tonally wrong content. Your job is to catch that and fix it before it goes live.
Add what only you can add. A specific client story, anonymised appropriately. A counterintuitive observation from your sector. A direct answer to the question your buyers always ask but most content in your space never actually addresses. These are the elements that make your content irreplaceable — and these are the elements AI cannot generate for you.
Build the validation layer that AI cannot fake. Great content is necessary but not sufficient. The reviews, the press mentions, the case studies, the consistent entity signals — these are what tell Google and AI platforms that your content is backed by a real, trustworthy business. Build those signals deliberately alongside your content strategy.
Google’s May 2026 core update is not anti-AI. It is anti-hollow. It is anti-generic. It is anti-content-that-was-made-for-algorithms-instead-of-people.
If you have been using AI as a tool to help you create more genuinely useful content faster — you are fine. Keep going. Refine the process. Make it more human, not less.
If you have been using AI to generate content at scale without genuine expertise behind it — the update you just experienced is a diagnostic, not a death sentence. Fix the foundation. Rebuild with real substance. Add the validation signals that prove your expertise is real.
The mixer does not make the baker irrelevant.
It makes a good baker exceptional.
Be the baker.
DIY — The AI Translator
My book The AI Translator — co-authored with Forbes contributor Michael Ashley and available on Amazon — gives you the complete 4 Vs framework in plain language, with step-by-step guidance on building the content strategy and validation architecture that wins on both Google and AI platforms simultaneously.
Get The AI Translator on Amazon →
Done With You — Iffel International
If you want Iffel International to audit your content signals, diagnose your 4 Vs gaps, and build the GEO, AEO, and Agentic Engine Optimization infrastructure that turns your content into AI-cited, Google-rewarded, lead-generating authority — let’s talk.
Call Emma, our AI concierge, at (949) 779-6442 for a complimentary consultation.
Hema Dey is the Founder and CEO of Iffel International Inc., a Forbes Top 5 AI Leader, and co-author of The AI Translator with Forbes contributor Michael Ashley. She is known as The AI Translator and the Chindian CEO. Iffel International helps businesses across law, healthcare, manufacturing, and professional services build AI-powered visibility and revenue systems through SEO2Sales™, GEO, Signal2Phygital™, and Agentic Engine Optimization. Based in Anaheim Hills, Orange County, California.
Your voice comes from your thinking — not your typing. AI cannot replicate the insight you developed from a difficult client conversation. It cannot reproduce the counterintuitive observation you formed from twenty years in your sector. It cannot generate the specific example that only you lived through.
Start every piece of content with three things only you know. Then invite AI in to help you structure and express them. Your thinking first. AI’s efficiency second. That sequence protects your voice every time.
Not only is it okay — it is the correct way to use it. The overall direction and final editing decisions should always come from a human. When you make AI your assistant — not the last word — you will have something much better at the end.
AI has a habit of coming up with lines filled with adverbs, phrases not used in spoken language, and long wordy descriptions without specific examples or further proof. These parts need to be edited manually and replaced with more in-depth, specific, human information. Financer
The fix is simple but requires discipline. Read every sentence out loud. If you would not say it in a conversation with a real client — delete it and rewrite it in the language you actually use. Add one specific example from your real experience to every major point. That one habit eliminates most AI-sounding content instantly.
Google does not penalize AI-assisted content. The debate is no longer about whether AI can write — it is about how we make good use of these tools. The concern is accuracy, authenticity, ethics, and over-reliance on automation. Iffelinternational
The rule is simple: if it genuinely helps the person reading it, Google rewards it. If it was produced for volume without real expertise behind it — Google finds it eventually.
Use AI for: ideas, structure, first drafts, grammar, simplification, alternative phrasing, and gap checking.
Never use AI for: your point of view, your client stories, your sector-specific insight, your claims that need verification, or your conclusion. Those must come from you — because those are the things Google and AI platforms reward and that your readers actually remember.
Call Us: 949-779-6442
If you are falling in love with us,
learn our love language with this eBook
before we seal the deal...
