What CMS Is Winning at AI in 2026?
Every CMS vendor now sells AI. None of them built the model behind it. The race is being won on what surrounds the model, and most buyers cannot yet tell.
The short answer
No CMS is winning at AI by building AI. Every platform here wraps models from OpenAI, Anthropic or Google, so the race is decided by what surrounds the model: agents with approval gates, an MCP server, a choice of model and an audit trail. Contentstack, SitecoreAI and Contentful lead on that measure. Optimizely SaaS CMS and Adobe Experience Manager follow for enterprises, Sanity for teams building their own agents, and HubSpot Content Hub for marketers without developers.
Key takeaways
- No CMS vendor trains a frontier language model. SitecoreAI runs on Microsoft's Azure OpenAI service, Optimizely's Opal on Google's Gemini and Anthropic's Claude, HubSpot's model cards name OpenAI, Anthropic and Google feature by feature, and Adobe's only home-grown generative model is Firefly, for images. What you are buying is the wrapper.
- There are two strategies. Wrap a lab's model into named agents and meter it in credits, or open a door with an MCP server and bring-your-own-model so your own agents and assistants do the work. The platforms at the top of this editorial do both, and that is why they are at the top.
- Text generation and MCP access are now table stakes. Across the 41 platforms we score, the thin spots are semantic search, machine-learning personalization, AI audit trails and usage dashboards. That is exactly where the surprise bills and the compliance problems appear.
- Agentic is the most abused word in CMS marketing this year. Several vendors' agent suites are early access, preview or a vendor-run service dressed as a product. Buy what is generally available, with an approval step you can see and a log you can export.
- Marketers are winning where the agent sits inside the editor with brand rules and an approval gate. Developers are winning where the door is open. Legal is winning only where the vendor names its model provider in writing and bars training on your data.
Our picks at a glance
ContentstackPick the model per agent, MCP for everything else.
SitecoreAITwenty agents, certified governance, one vendor's model.
ContentfulFive providers on your keys, audited.
SanityTwo MCP servers, embeddings, published credit price.
HubSpot Content HubBrand-voice agents at published prices.
DrupalForty-plus providers, your keys, your servers.
Our analysts’ take
Walk any CMS vendor's website in October 2026 and the word AI appears before the word content. SitecoreAI renamed the whole product around it. Kontent.ai put it in the company name. Optimizely, Adobe, Contentstack and HubSpot each sell an agent platform, and even the open-source projects ship an AI foundation in core. A marketing leader is told, repeatedly, that the platform will write, translate, audit, personalize and publish on its own. What nobody says on the home page is where the intelligence comes from.
The answer is the same everywhere. It comes from OpenAI, Anthropic or Google, rented by the token and resold inside the product. That is not a scandal; it is the only sensible way to build software in 2026. But it changes the question a buyer should ask. If every vendor has access to the same handful of models, the model is not the differentiator. The differentiator is what the vendor built around it: whether the agents can do real work in your content, whether a human approves before anything publishes, whether you can see what it cost, whether you can swap the model when your legal team or your budget says so, and whether your own AI tools can reach the content through a standard door.
This editorial takes a position on that question. It names the two strategies vendors have chosen, says which platforms are executing each one and which are mostly announcing, and reads the evidence across all 41 platforms we score to answer the question buyers actually have: is anyone winning yet, and is it you? This is where the Principal Analyst Group stands in October 2026.
What to look for
- 1
Whose model, really
Ask which lab's model runs each AI feature, whether the vendor can swap it without telling you, and whether you can bring your own keys. The answer decides your data agreements, your exit options and who profits from the markup.
- 2
Agents that do work and stop
A real agent takes a brief, builds or changes many pages, and queues the result as drafts for a human. A chat box that rewrites a paragraph is an assistant. Ask for the multi-step demo and for the screen where you reject the result.
- 3
A door your own tools can use
An official MCP server lets ChatGPT, Claude, Copilot or the agents you build read, write and publish under your permissions. Check whether it is hosted or local, whether it respects roles, and whether every action lands in the activity log.
- 4
Guardrails legal will accept
An audit trail that separates human from AI edits, brand and legal rules the AI cannot skip, a contract that bars training on your data, and ideally ISO 42001 certification of the vendor's AI management. Without these, the feature stays switched off.
- 5
A bill you can forecast
Every platform meters AI in credits, interactions or tokens. Ask for the credit price, the included allowance, the cap and the dashboard, and ask what a campaign launch with translations consumes. The license is rarely the expensive part.
The ratings at a glance
Ranked by Fit score for this topic, as rated on October 3, 2026. Dots run from one (Limited) to five (Excellent); each review below spells the ratings out.
| Platform | Fit | AI | Compliance | Self-service | Growth tools | Budget | Upkeep |
|---|---|---|---|---|---|---|---|
| 1Contentstack | 7.7/10 | Strong | Strong | Good | Good | Good | Good |
| 2SitecoreAI | 7.7/10 | Strong | Strong | Strong | Good | Limited | Fair |
| 3Contentful | 7.6/10 | Strong | Strong | Good | Fair | Fair | Fair |
4 Adobe Experience Manager | 7.3/10 | Strong | Strong | Strong | Strong | Limited | Limited |
5 Optimizely SaaS CMS | 7.3/10 | Strong | Good | Good | Good | Limited | Good |
| 6HubSpot Content Hub | 7.1/10 | Good | Strong | Good | Strong | Good | Fair |
| 7Sanity | 7.1/10 | Good | Good | Good | Limited | Good | Fair |
8 Kontent.ai | 7.0/10 | Good | Strong | Good | Limited | Good | Good |
9 Webflow | 6.8/10 | Good | Good | Strong | Good | Fair | Fair |
| 10Drupal | 6.0/10 | Fair | Fair | Good | Fair | Strong | Limited |
- 1Contentstack7.7/10
- AI for marketers
- Strong
- Security & compliance
- Strong
- Marketer self-service
- Good
- Campaign & growth tools
- Good
- Budget predictability
- Good
- Low IT upkeep
- Good
- 2SitecoreAI7.7/10
- AI for marketers
- Strong
- Security & compliance
- Strong
- Marketer self-service
- Strong
- Campaign & growth tools
- Good
- Budget predictability
- Limited
- Low IT upkeep
- Fair
- 3Contentful7.6/10
- AI for marketers
- Strong
- Security & compliance
- Strong
- Marketer self-service
- Good
- Campaign & growth tools
- Fair
- Budget predictability
- Fair
- Low IT upkeep
- Fair
- 4Adobe Experience Manager7.3/10
- AI for marketers
- Strong
- Security & compliance
- Strong
- Marketer self-service
- Strong
- Campaign & growth tools
- Strong
- Budget predictability
- Limited
- Low IT upkeep
- Limited
- 5Optimizely SaaS CMS7.3/10
- AI for marketers
- Strong
- Security & compliance
- Good
- Marketer self-service
- Good
- Campaign & growth tools
- Good
- Budget predictability
- Limited
- Low IT upkeep
- Good
- 6HubSpot Content Hub7.1/10
- AI for marketers
- Good
- Security & compliance
- Strong
- Marketer self-service
- Good
- Campaign & growth tools
- Strong
- Budget predictability
- Good
- Low IT upkeep
- Fair
- 7Sanity7.1/10
- AI for marketers
- Good
- Security & compliance
- Good
- Marketer self-service
- Good
- Campaign & growth tools
- Limited
- Budget predictability
- Good
- Low IT upkeep
- Fair
- 8Kontent.ai7.0/10
- AI for marketers
- Good
- Security & compliance
- Strong
- Marketer self-service
- Good
- Campaign & growth tools
- Limited
- Budget predictability
- Good
- Low IT upkeep
- Good
- 9Webflow6.8/10
- AI for marketers
- Good
- Security & compliance
- Good
- Marketer self-service
- Strong
- Campaign & growth tools
- Good
- Budget predictability
- Fair
- Low IT upkeep
- Fair
- 10Drupal6.0/10
- AI for marketers
- Fair
- Security & compliance
- Fair
- Marketer self-service
- Good
- Campaign & growth tools
- Fair
- Budget predictability
- Strong
- Low IT upkeep
- Limited
The shortlist, reviewed
Contentstack
The clearest case of doing both. Agent OS ships the Polaris assistant, a no-code agent builder and automations, and since October 2026 Polaris drafts the agent for you. You choose the model behind each agent across OpenAI, Azure, AWS Bedrock and Google, every run logs its model and tokens, and an MCP server with about a hundred tools opens the door.
- Best for
- Enterprise content teams on a modern web stack that want governed agents today and the right to decide which lab's model runs them.
- Watch out for
- The MCP server is still installed locally rather than hosted, semantic search covers uploaded documents in the Knowledge Vault rather than all content, and self-hosted or fine-tuned model endpoints are not a supported configuration.
- What it costs
- Free plan, Build at $29 per month, Growth at $299 per month plus $25 per extra user; Enterprise quote-only. AI runs on metered credits with configurable limits.
SitecoreAI
The best-executed wrapper in the market. Agentic Studio ships 20 named agents and a visual canvas for more, the Marketer MCP server lets Claude, Copilot Studio and Cursor run pages and experiments, and the AI management system is certified to ISO 42001. All of it runs on Microsoft's Azure OpenAI service, and that is the catch.
- Best for
- Enterprises that want agents, personalization, customer data and AI search visibility from one vendor and have a partner to run it.
- Watch out for
- You cannot bring your own model for anything except image tagging, approvals happen in chat rather than a policy you can set per agent, pricing is quote-only, and AI use is capped by a fair-use limit of 50,000 interactions a year per organization.
- What it costs
- Quote-only. No published tiers or free plan. AI usage is subject to a 50,000 interactions per year fair-use threshold.
Contentful
The widest-open door in headless. Five model connectors, from OpenAI and Claude on AWS Bedrock to Gemini, Azure OpenAI and any compatible endpoint, run AI Actions on your own keys. The hosted MCP server went generally available in July 2026 with OAuth, and audit logs that separate human from AI edits stream to your security tools within minutes.
- Best for
- Enterprises whose security review is the hard part, and developer teams that want AI to run under model contracts they already negotiated.
- Watch out for
- There is one named agent, the Optimization Agent, and it covers personalization and analytics rather than content production. Salesforce completed its acquisition in September 2026, and the roadmap for AI inside Contentful is now partly an Agentforce question.
- What it costs
- Free plan, Lite at $300 per month; Premium and Enterprise quote-only. AI Actions are metered, with usage reporting and a public Usage API.
Adobe Experience Manager
The one vendor that trains a generative model of its own, Firefly, and it makes images. Everything else runs on Adobe-managed models, with Microsoft's Azure OpenAI service on Adobe's sub-processor list. Named agents are in production through an Agentic add-on, the unified MCP server covers content, governance checks and deployment pipelines, and Content Credentials give AI output provenance.
- Best for
- Large enterprises standardized on Adobe that want agents, asset AI and personalization from one vendor and can fund a six-figure program.
- Watch out for
- Agents are a separate SKU on top of the most expensive platform here, bring-your-own-model is limited to translation through Azure OpenAI, agents are unavailable on the self-hosted 6.5 release, and consumption licensing counts every content request, including AI crawlers.
- What it costs
- Quote-only. Agents are an additional Agentic SKU. Implementations commonly run $500,000 to more than $1 million in services.
Optimizely SaaS CMS
Opal began as a wrapper around Gemini and became the largest agent catalog in the market, with more than 45 prebuilt agents and nested workflows behind them. Since March 2026 it also routes to Anthropic's Claude through Google Vertex, and Optimizely reserves the right to swap models. A hosted CMS MCP server and a per-agent credit dashboard round it out.
- Best for
- Marketing teams that run experiments and want AI search optimization and personalization handled by agents inside the editor they already use.
- Watch out for
- Model routing is Optimizely's choice, not yours, and customer-supplied keys are not a self-service option in the SaaS CMS. Everything is quote-only, AI is an add-on on credits, and heavy use is reported to add 20 to 40 percent to a contract.
- What it costs
- Quote-only. Priced on traffic and feature tier, with experimentation and AI as add-ons on credits.
HubSpot Content Hub
The wrapper marketers can actually use. Breeze agents draft posts, pages and emails in a brand voice trained on your writing samples, every agent action shows up as an audit card, and two official MCP servers connect outside assistants. HubSpot calls itself the Switzerland of AI: its model cards assign OpenAI, Anthropic or Google feature by feature.
- Best for
- B2B marketing teams on the HubSpot CRM that want AI content, AI search reporting and personalization at a published price with no developer.
- Watch out for
- There is no semantic search or vector layer for your own agents to build on, the remote MCP server is read-only for pages and blog posts, bring-your-own-key is limited to workflow actions on Enterprise, and the jump from Starter to Professional is steep.
- What it costs
- Published. Starter from $15 per month, Professional from $450 per month (three seats), Enterprise from $1,500 per month. AI runs on HubSpot Credits at $10 per 1,000 with enforceable caps; the Content Agent charges 1,000 credits per finished piece.
Sanity
The platform that treats the door as the product. Two hosted MCP servers, embeddings on every plan including Free, a public Content Agent API and guides for the OpenAI Agents SDK and LangChain make it the easiest content store to put behind agents you build. Content Agent stages every edit as a reviewable draft, and credits cost five cents each.
- Best for
- Developer-led teams and digital products that want to build their own AI workflows on a structured content store with transparent pricing.
- Watch out for
- The editor-facing AI runs on a blend of OpenAI, Anthropic and Google models that Sanity picks, with no provider choice; only the Context tooling is bring-your-own-key. Custom instructions are guidance rather than enforcement, there is no personalization engine, and nothing in the documentation indemnifies AI output.
- What it costs
- Free plan, Growth at $15 per seat per month (viewers free), Enterprise quote-only. AI credits are published at $0.05 each with spending limits.
Kontent.ai
The best paperwork in the race and the most closed model choice. Kontent.ai is certified to ISO 42001 across the whole platform, its Expert Agents for SEO, translation and compliance run at review checkpoints with every action logged and reversible, and nothing publishes without a person. Anthropic's Claude does the work, on zero-retention endpoints, and you cannot change it.
- Best for
- Regulated organizations and localization-heavy teams that need an audit trail a compliance officer will sign before they will accept an agent at all.
- Watch out for
- No bring-your-own-model of any kind, no runtime personalization engine, semantic search is for editors rather than developers, and pricing is a calculator that ends at a sales call. Marketing features beyond content are thin.
- What it costs
- Quote-only beyond a free Developer plan. Pricing is calculated on seats, content types, items, languages and environments.
Webflow
A wrapper for the editor and a wide door for everything else. Webflow partners with OpenAI and Anthropic and picks the model for you, but MCP 2.0 lets agents build across pages and sites on isolated branches, under your roles, with every action attributed to AI in the activity log. Webflow says more than 30 percent of enterprise customers use it, almost all through Claude.
- Best for
- Design-led marketing teams that own the website end to end and want AI search optimization and agent access handled where the pages are built.
- Watch out for
- You cannot bring your own model or keys, semantic search is absent, structured content is shallow next to the headless platforms, and Source, the governed agent workspace announced in September 2026, is a limited preview for select partners.
- What it costs
- Published. Site plans from $15 per month, Premium at $25 per month (annual), seats from $15 to $39 per month. AI features draw on credits included with each plan, with quotas enforced since June 2026.
Drupal
The purest door in the market. The Drupal AI module bundles no model: you supply keys for any of more than 40 providers, from OpenAI and Anthropic to a model on your own server, and pick a different one per task. More than 30 agents build content types and pages with human oversight, and usage exports to Datadog or Grafana.
- Best for
- Organizations with Drupal developers that want AI under their own data residency rules and their own provider contracts, with nothing resold.
- Watch out for
- There is no first-party agent product doing marketing work end to end, the MCP server is community-maintained and the newer one has no stable release, and there is no personalization engine. Everything takes a developer to assemble.
- What it costs
- Free software. You pay model providers directly at their list prices, plus hosting and whoever assembles and maintains the modules.
Two ways to put AI in a CMS, and nobody owns the model
Strip away the branding and every CMS vendor has made one of two bets, or both. Neither involves training a model. Adobe's Firefly is the lone exception, and it generates pictures, not pages. Everything that writes, translates, tags or reasons in these products is rented from OpenAI, Anthropic or Google, usually through Microsoft Azure, Amazon Bedrock or Google Vertex, and resold to you in credits.
The wrapper: pick a lab, build agents, meter it
The vendor chooses the model, builds brand kits, agents and approval flows around it, and sells usage back in credits or interactions. SitecoreAI on Azure OpenAI, Optimizely's Opal on Gemini and Claude, HubSpot's Breeze on OpenAI, Anthropic or Google depending on the feature, Webflow with OpenAI and Anthropic, Xperience by Kentico and Sitefinity on Azure OpenAI, Hygraph on Claude alone, Bloomreach's marketing agent on Gemini. The upside is finished features a marketer can use on day one. The downside is that the vendor owns the model choice, the markup and the roadmap, and swaps the model when it suits them.
The door: an MCP server and your own keys
The vendor exposes content, schema, workflow and publishing through a Model Context Protocol server and lets you bring your own model keys, then gets out of the way. Drupal, WordPress 7.0, Umbraco, Directus, Strapi, Magnolia and Jahia all work this way, and so does Sanity for teams building agents. The upside is control: your contracts, your data residency, your choice of Claude, ChatGPT or Copilot. The downside is that the vendor has shipped plumbing, and someone on your side has to build the house.
Both: the platforms actually winning
Contentstack lets you choose the model behind each agent and ships an MCP server with about a hundred tools. Contentful runs its AI Actions on five providers with your keys and hosts an MCP server with OAuth. Sanity builds its own agent on managed models and opens two MCP servers beside it. These are the platforms at the top of this editorial, and the lesson is not subtle: the vendors with the most finished AI are also the ones least interested in locking you into their model.
Is anyone winning yet? What the evidence across 41 platforms says
Read across every platform we score and the picture is less flattering than the home pages. Fewer than one platform in eight has built-in AI we would call strong across the board. More than a third sit in the bottom half, with an AI button at most. And the strongest AI belongs, almost without exception, to the platforms that were already the strongest content platforms before AI arrived. AI has produced no upsets. It has widened the gap between the large vendors and the small ones, and between the platforms with structured content and the ones storing pages as HTML.
What is solved: writing and the door
In-editor text generation with some brand control is now the norm, and an official or well-supported MCP server is the single most common AI feature in the market: nearly three quarters of the platforms we score ship one. Several platforms with almost no AI inside the editor, among them Strapi, Agility CMS, DatoCMS and Ghost, have excellent MCP servers. The door is open everywhere. What is behind it varies enormously.
What is thin: search, personalization, audit, cost
Semantic search your own applications can query exists on about five platforms. A real machine-learning personalization engine exists on about eight, nearly all of them enterprise suites. AI usage dashboards with cost tracking are rarer still, and a full audit trail of who invoked AI and what changed is confined to the same handful. These are the four capabilities that decide whether AI pays for itself and whether legal lets you keep it on, and most of the market has not built them.
What is announced: the vaporware list
Acquia AI is early access with no service commitment, one sandbox agent and no production use. Squiz Spark is recruiting beta testers and appears in no release note. Jahia's eighteen agents are a vendor-run migration service, not a product you operate. Every agent in Xperience by Kentico's marketing suite is flagged as a preview feature. Source by Webflow is a limited preview. Payload's AI workflows are a beta. HCL Digital Experience, Concrete CMS and Kentico Xperience 13 have no agents and no MCP server at all. Gartner coined the term agent washing for this in June 2025, predicted that more than 40 percent of agentic AI projects would be cancelled by the end of 2027, and found in October 2025 that 45 percent of marketing technology leaders running vendor agents said they fell short of the promised results.
Who is winning: marketers in the suites, developers at the door
Marketers win where the agent sits inside the editor with brand rules and a visible approval step: SitecoreAI, Optimizely SaaS CMS, Contentstack and HubSpot Content Hub do this today, and Adobe does it for the teams that can fund it. Developers win where the door is open and the pricing is published: Sanity, Contentful, Drupal, Directus and Strapi. Nobody wins where the vendor has wrapped a model, hidden its name, metered it and shipped no approval gate. Several platforms in the middle of the market are exactly there.
How to tell bleeding edge from buzzword before you sign
Ask which model, in writing
Ask for the name of the lab and the cloud behind each AI feature, whether the vendor can change it without notice, and whether the contract bars training on your content. A vendor that will not name the provider is reselling something it does not want you to price. Contentstack, Contentful, HubSpot, Sanity and Kontent.ai publish the answer. Strapi, Prismic, Squiz and Uniform's Scout agent do not.
Ask for generally available, not announced
Agent suites move from keynote to preview to early access to general availability, and the pricing page rarely says which stage you are buying. Ask for the release note and the date. If the demo environment is the vendor's and not yours, the feature is not shipping.
Watch the agent get rejected
Have the agent build a campaign or translate a section, then ask to see the screen where an editor rejects it and the log entry that recorded who asked, what changed and which model ran. If the answer is a chat transcript, the approval gate is a conversation, not a control.
Price a month of real use
Get the credit price, the included allowance, the overage rate and the cap, then ask what a product launch with twelve translations consumes. Sanity publishes five cents a credit; SitecoreAI caps interactions at 50,000 a year; Optimizely sells AI as an add-on that can add 20 to 40 percent. The platform that looks cheapest on the license is often not.
Try the door with your own assistant
Connect Claude, ChatGPT or Copilot to the vendor's MCP server using an ordinary editor's account and ask it to draft, change and publish something. You will learn in ten minutes whether permissions hold, whether the action lands in the audit log, and whether the vendor's door is a product or a press release.
Frequently asked questions
Which CMS is winning at AI in 2026?
Contentstack, SitecoreAI and Contentful lead among the platforms we score, because each pairs working agents with governance and, in Contentstack's and Contentful's case, lets you choose the model. Optimizely SaaS CMS and Adobe Experience Manager follow for enterprises, Sanity leads for teams building their own agents, and HubSpot Content Hub is the strongest choice for marketing teams without developers.
Does any CMS vendor have its own AI model?
No CMS vendor trains a frontier language model. Every platform's writing, translation and agent features run on models from OpenAI, Anthropic or Google, usually through Microsoft Azure, Amazon Bedrock or Google Vertex. Adobe's Firefly is the one home-grown generative model of note, and it generates images. Bloomreach and Adobe also run their own machine-learning models for personalization, which is a different thing from generative AI.
Which AI models do SitecoreAI, Optimizely and Adobe actually use?
SitecoreAI's Stream and Agentic Studio run on Microsoft's Azure OpenAI service. Optimizely's Opal was built on Google's Gemini and since March 2026 also routes to Anthropic's Claude through Google Vertex, with Optimizely choosing the model. Adobe runs its agents and assistants on Adobe-managed models, with Microsoft's Azure OpenAI service named on its sub-processor list, uses its own Firefly model for images, and lets you connect your own Azure OpenAI subscription only for translation.
Should I choose a CMS with built-in AI agents or one with an MCP server?
If your marketing team will use AI daily without a developer, choose a platform with finished agents, brand rules and an approval gate inside the editor, such as SitecoreAI, Contentstack, Optimizely SaaS CMS or HubSpot Content Hub. If you have developers and want AI under your own model contracts, an open door matters more: Contentful, Sanity, Drupal, Directus and Strapi. The best platforms offer both, and an MCP server is worth insisting on either way.
Is AI in a CMS worth paying extra for in 2026?
Yes when the platform has agents that do multi-step work with an approval step and a usage dashboard, because that saves real editorial hours and the cost is visible. No when the AI is a rewrite button metered in credits with no cap. Ask for the credit price and the included allowance before signing. Sanity publishes $0.05 per credit, HubSpot and Webflow include credits with their plans, SitecoreAI caps use at 50,000 interactions a year, and Optimizely's AI add-on can raise a contract by 20 to 40 percent.
How we rated these platforms
Each rating combines the relevant criteria from our independent scorecards into a five-point scale. The Fit score blends the ratings below, weighted for this topic, into one number out of 10. The ratings were taken from the scorecards on October 3, 2026 and are fixed with this editorial; follow the scorecard links above for the current ratings.
Scorecards and platform inclusion are unsponsored: no vendor pays to appear in this guide or to change a rating. Full criteria, weights and evidence are in the methodology and on each platform’s scorecard. Spotted something out of date? Tell us.
Still deciding?
Answer seven questions for a shortlist built around your team, or put the top two side by side.