Did you see the viral AI ad aired at the NBA finals? Crazy visuals, impeccable product integrations, 3.3M views on X. I almost couldn’t believe it was completely AI-generated content.

Get Started with HubSpot's AI Campaign Assistant

An AI film-maker co-wrote the script with Gemini, generated Veo-3 prompts and videos, and edited clips on Capcut, all with a budget of just $2,000.

This ad made me realize we’ve hit a tipping point in AI-generated content. If you’re a content marketer, you need to pay attention to AI now.

In this blog post, I’ll dive deeper into the top AI-generated marketing content types, using data from HubSpot’s State of AI survey, which includes insights from over 1,500 U.S. marketers. I’ll also share AI content examples and tips so you can see the potential in action.

Table of Contents

Marketers, AI-Generated Content, and Today’s AI Landscape

According to HubSpot’s State of Marketing 2025 survey, marketers think it’s time to go all in on AI.

And I completely agree.

First off, AI adoption remains high but still has room for growth, particularly in non-US markets. According to HubSpot’s 2025 State of AI Marketing Report, approximately two-thirds of marketers indicated AI use at work.

While this is lower than last year’s figure (74%), that can be accounted for by the difference in sample constitution, as this year surveyed a global audience whereas last year only surveyed U.S. respondents. The U.S. number has held steady since last year.

On the tech front, we’re seeing improvement on multiple fronts. Models like GPT-4 and Claude 3.5 now have great reasoning and writing capabilities.

AI agents are also becoming more powerful. Technologies like the Model Context Protocol (MCP), which allow LLMs and agents to share information with each other, are pushing us towards the agentic era.

In the context of marketing, AI has seen widespread adoption and practical wins. The vast majority (89%) of AI users indicated that their AI usage had increased as a result of it being added to their existing tools.

Media content creation (38%) is an emerging use case, indicating AI’s growing role in visual and audiovisual content, though it still lags behind text. Brainstorming (34%), skill learning (33%), and data analysis (36%) are additional applications, though less central to daily marketing workflows.

Administrative AI applications, like meeting note-taking (26%) and internal communications (14%), remain niche, reinforcing that AI is primarily valued for content production rather than internal operations.

But it’s not all smooth sailing.

AI doesn’t have the best reputation when it comes to accuracy and biases. AI-powered recruiting can have built-in age-discrimination Sometimes, it even tells business owners to break the law.

The regulatory landscape is shifting to attempt to mitigate AI risks and keep up with the pace of innovation.

At the time of writing this article, over 70 countries have their own AI policies and frameworks to promote responsible AI development. The European Union’s proposed AI Act aims to regulate high-risk AI applications, including those in marketing, to ensure ethical practices and protect consumers.

This regulatory push is vital for maintaining trust and preventing AI misuse.

Most marketers I’ve spoken to share a similar sentiment: excited about everything they can do with AI, but nervous about the potential impacts on their jobs and the constant pressure to upskill.

Personally, as someone who has always been drawn to automation, I find AI-generated content fascinating. At the same time, I can’t ignore the impact it will have on society as a whole.

Let’s look at the pros and cons of AI-generated content marketing.

AI-Generated Content in Marketing: Pros and Cons

Pro: Efficiency and Cost-Effectiveness

With AI, teams have become leaner. One person with the right AI model and prompt can recreate work that previously needed a creative team and weeks of effort. 55% of the marketers we surveyed cite time savings as the key reason why they use AI.

Additionally, 65% of marketers say that AI allows them to focus on more creative aspects of their job by freeing up their time. Meanwhile, companies are actively looking for areas to automate so resources can be allocated more efficiently.

For instance, Shopify CEO Tobias Lütke has an aggressive approach to drive AI adoption. His internal memo states, “Before asking for more headcount and resources, teams must demonstrate why they cannot get what they want done using AI.”

Companies are also building out their own “AI team members.. Over 66% of marketers plan to build out internal tools for their marketing teams, expecting a 45% cost reduction to external tools.

Simply put, AI allows teams to scale without increasing headcount.

ai-generated content in marketing: pros and cons

Con: Quality and Authenticity Concerns

AI-generated content often gets a side-eye for being too robotic or making up facts. Based on HubSpot’s AI survey, marketers continue to have serious concerns about generative AI:

  • 43% say AI sometimes produces inaccurate information.
  • 34% say AI can get biased.
  • 30% believe that AI produces vague and surface-level content.

Of course, your experience might be different based on the model you use, how you’re prompting it, and the subject matter itself. AI might do okay while generating an email for a team update, but it may make up facts while writing about the latest advancements in data science.

I’ve found it often confuses software versions, report statistics, and references outdated information. Video generators like Veo-3 also frequently produce misspelled subtitles.

As HubSpot channel monetization lead Alexe Cunningham mentions, “AI is there to help us but isn’t the source of truth. It’s a jump start on whatever we are doing. We always check our work.”

Pro: Personalization

AI videos as cold DMs. Dynamic content delivery based on past preferences. Personalized real-time recommendation systems.

Sound like a lot of work? This level of personalization would be impossible to achieve manually (especially if you have thousands of visitors). However, AI can personalize all of this and more at scale.

For instance, HubSpot’s demand gen team used AI to comb through thousands of customers and send personalized content recommendations to them. One result was a cold brew company that had downloaded influencer marketing resources and also showed interest in content planning for seasonal launches.

Instead of generic resources, HubSpot’s AI system recommended a content marketing course with a targeted copy. Here’s what they sent: “Turn every sip into a story that captivates and converts.”

Using AI can help your brand stop the scroll and deliver an engaging and intimate experience to your customers.

Con: Ethical and Privacy Issues

AI systems are trained on loads of data, some of which are sourced quite questionably. I’ve seen lawsuits on models scraping private user data without consent. Artists have also filed lawsuits against AI companies for training on copyrighted artwork.

In a world where AI can create convincing deepfakes and fabricated case studies, it’s becoming difficult for customers to know what’s actually real.

Countries are investing heavily in a broader AI strategy to help combat ethics and privacy issues. Along with GDPR, regulations like the European Union’s proposed AI Act and California’s Consumer Privacy Act (CCPA) impose strict data privacy rules. Data marketplaces like Kaggle help provide legal and standardized data, but the battle of privacy and ethics with AI is far from over.

Pro: Analysis and Research Capabilities

With the release of models like OpenAI’s deep research and GPT-4, AI has become powerful in research and data analysis.

I recently tested some AI tools for Excel for another article and was pretty impressed at how ChatGPT processed complex datasets. I uploaded months of YouTube data, and it was able to identify content patterns and formats that I had missed. These insights were immensely helpful for my content strategy going forward.

Similarly, AI can help with analyzing and tweaking content performance across campaigns.

Con: Greater Dependence on Technology

I came across an interesting study by MIT. Researchers had students write SAT essays. Some of them were allowed to use ChatGPT, while others weren’t. The results are honestly a bit scary, given how every marketer is learning AI right now.

Students using ChatGPT showed the lowest brain engagement and barely remembered what they wrote shortly after. By the third essay, they had given up completely and simply copied and pasted what ChatGPT provided.

This shows how AI can become a crutch, even when you don’t want it to.

So, does that mean you should completely write off AI and do everything manually? Definitely not.

The study also showed that students who first wrote the essays themselves and then used AI to enhance their writing performed better than both the previous groups. The takeaway here is: use AI to amplify your creative process, not replace it.

It’s essential to be mindful of your AI usage and consider setting limits. I prefer to work on a rough draft or content strategy before I head to AI to polish it. You can choose to complete some projects without using AI at all.

The Top 3 Content Types for Generative AI

According to HubSpot’s 2025 State of AI Marketing Report, marketers are using AI most heavily for the following three use cases.

1. Emails and Newsletters (51%)

I personally find emails the most challenging content format to get right. You’re competing for attention against hundreds of emails an average person receives daily. One wrong word, and you’re either in the spam or promotions folder, killing your open rates.

Even when you make it to the primary inbox, you’ve seconds to convince them to open your email and take action. This is why 51% of marketers use AI to help them with email and newsletter content.

You can use AI to brainstorm campaigns, personalize emails, and A/B test results. Here’s how.

  • Email strategy. Instead of guessing what your customers want, AI can help you figure out what they’re actually engaging with. Feed AI your ideal customer profile and ask it to reverse engineer the kind of messaging and subject lines they’re most likely to click on. Use that information to inspire your email strategy. I also use AI to brainstorm lead magnet outlines and email sequences.
  • Personalization. Traditionally, email personalization is done in batches. You segment your audience by a specific factor or action and then swap names, job title, or any other details they’ve filled in a form. But with AI, you can actually drive one-to-one personalization at scale. For instance, HubSpot’s demand generation team used AI to increase conversions by 82% through analyzing user behavior and website data.
  • A/B testing. I’ve always found A/B testing a bit hard, whether it’s for emails or social media. You have to write out variations of the same email, adjust for multiple variables, and then dig into the data. It’s really easy to get stuck in analysis paralysis since it’s a long and tedious process. Good news? AI can tackle a lot of the heavy lifting here. Use AI to do real-time data analysis and personalize, continuously refine A/B tests, and measure the results.

Pro tip: Tools like HubSpot’s AI Email Writer help me generate multiple variations of my emails for efficient A/B testing.

ai-generated content, hubspot’s ai email writer

Source

2. Social Media Content (49% for Text-Based, 47% for Video/Audio)

Coming in a close second, 49% of marketers use AI for text-based social media content, and 47% for video/audio generation. Honestly, I’m surprised it’s not the first.

Since I work extensively in social media, I’m noticing a clear trend toward AI-generated content across industries. Apart from the AI ad I mentioned earlier, my feed is filled with influencers using AI-generated avatars to create videos faster.

Personally, I use AI as a content assistant. For instance, one of my clients has built a CustomGPT trained on years of his posts. So, instead of writing his content from scratch, my job is to feed unique research into the GPT and then refine the output for flow and accuracy.

Here’s how other marketers are using AI for social media.

  • Content creation. AI tools can help generate content ideas, conduct heavy research, create first drafts, and even generate realistic art and video. I’d recommend checking out the following tools: ChatGPT (analysis), Sora (image generation), Claude (writing), Veo-3 (video creation), and HeyGen (AI avatars).
  • Campaign automation. Posting and scheduling can be a tedious task, especially when you’ve got a lot of platforms to keep active. AI scheduling tools can completely take this off your plate. I use AI to automate posting across platforms, with optimized content for each. n8n, a workflow automation software, is also popular for repurposing content and posting on different platforms.
  • Analytics. Use AI on your social media analytics to spot patterns between content formats, types, and posting times. You can also analyze external datasets to gain insight into viral hooks and social trends. Once you understand what’s working, you can feed this data into your content creation workflow and double down on it.

I spoke with Spencer Tahil, an AI consultant at Growth Alliance, and thought his content creation process was quite interesting. He turns meeting transcripts into on-brand LinkedIn posts using a series of AI prompts. These prompts create a voice profile by identifying unique communication patterns, linguistic preferences, and then calibrating the final output.

“[Using this system], I get content that actually sounds like me — not some corporate robot. I’m getting 95% usable content on the first draft, saving me hours every week,” he told me.

3. Long-Form Content — Blogs, Articles, etc. (46%)

Quality long-form content can do a lot for your business: drive sales-qualified leads, build authority, and nurture your leads. However, it also can take forever to write.

That’s why 46% of marketers (and me!) are actively using AI for writing blog posts and articles. Here are some ways you can use AI writing generators for long-form content.

  • Efficiency. AI can drive efficiency gains in almost every part of the long-form creation process. What I love most is that I don’t have to stare at a blank page to write a blog post. Marketers can use AI to generate blog outlines, do research, and even generate first drafts. I also use it to spot any logical or reasoning gaps in my pieces and tighten up the copy.
  • Content optimization. With AI, you can easily analyze what type of content is working in your favor. For instance, HubSpot’s analytics revealed that our readers prefer blog posts with case studies and real-life examples. So, I started adding more of these elements, and engagement shot up.

Daria Bulatovych, a content strategist at TripRobotics, told me how AI helped her cut down her blog creation time.

“A typical piece took a full workweek: 3-4 days of research, 1.5 days of writing, and additional time for edits and SEO optimization. However, I can now create equally strong content in just 1.5 days [with GPT4]. These articles drive consistent traffic in a highly competitive niche.”

Pro tip: Ready to use AI and speed up your long-form creation process? I’d suggest opting for HubSpot Breeze as your end-to-end content assistant. Breeze can ideate, create, and distribute content that speaks to your customers.

The best part? Breeze natively syncs with HubSpot’s CRM to pull up data and insights from your sales funnels. You get content strategies perfectly aligned to your marketing funnel without switching between multiple tools.

5 Tips for Using AI-Generated Content in Marketing

1. Treat AI as a teammate.

I like to treat AI as an intern. Just like you wouldn’t hand over your business to a new team member and expect them to fix everything, don’t expect AI to give you polished results without context.

I view AI as a collaborator to my business. That means I set up projects with onboarding documents: business goals, strategy documents, current initiatives, and more. Just like training an intern, I set clear expectations on how I want it to help me.

When you collaborate with AI, you’ll start to unlock efficiency gains. For instance, I use AI for editing first drafts for flow, brainstorming visuals for content pieces, and gaining insights from my data.

The best part? I can delegate assignments to AI 24/7, get revisions back in seconds, and prompt it to refine the work till I get what I need. But just like human teammates, you get what you put in.

2. Edit, edit, edit.

AI-generated drafts are just a starting point. Your final content should reflect your brand’s unique voice and insights, ensuring it stands out and offers real value.

Here are some elements I include to make my content unique and non-replicable:

  • Personal opinions on a topic.
  • Internal company data and product materials.
  • Quotes from SME interviews.
  • Actually testing products and tools.
  • Behind-the-scenes footage and stories of a launch.
  • Photos and screenshots
  • Rigorous fact-checking

So, even if you use AI to generate a first draft, make sure you personalize it with real-world experience.

3. Choose the right content tasks to automate.

AI has the potential to automate a lot of your content creation process. The question is: Should you let it?

There is no blanket answer because using AI is heavily dependent on your priorities, current processes, and goals.

For instance, I prefer not to use AI for my personal social content. Since I don’t have a library of content that performs well already, there is no context for AI to build on. It can’t replicate my voice (yet), because even I am not sure what my voice is.

Typically, I write first drafts myself and then ask AI to polish and tighten the copy. I also turn to AI for content ideas when I’m stuck.

On the other hand, marketers like Jess Cook from Vector have built a system around automating founder-led content at scale. I find her approach incredibly effective, and maybe in the future, it’ll make sense for me too.

Similarly, if you’re an SEO company, you’d want to build AI processes around keyword research automation or audits. In contrast, if you specialize in thought leadership, you’d want to automate research while keeping the insights human.

Resist the temptation to automate every content task, and focus on the ones that will actually move the needle for you.

4. Work on your prompts.

Good content starts with good prompts. If you want to create AI-generated content at scale, prompting is the first thing you should learn.

After testing and iterating over hundreds of prompts, I’ve learned that good prompts are clear and specific. The more context, examples and data you can give AI, the better output it’ll give you.

Here are four things I recommend you add to your prompts:

  • Role. Give AI a role to play. If you’re working on emails, ask it to act as an “email specialist with 10+ years of experience.” I also specify the industry I’m working in (e.g. B2B SaaS) to get even more nuanced responses.
  • Tone. Include the tone and formatting you want AI to respond in, e,g. conversational, fun, edgy, short, punchy, etc.
  • Context/Examples. My most successful prompts include real-world examples. Think: content which has performed well earlier, competitor examples, outlines, strategy documents, and more.
  • Constraints. It’s also important to specify what you want the output to not look like. Introduce guardrails to limit AI hallucinations. This could look something like “favor plain language over jargon, don’t make claims about ROI without data, avoid referencing X publication,” etc.

5. Keep up with AI.

AI is getting better every single day.

A few months ago, I had to generate AI video and audio separately and sync them separately. Today, I can just use Veo-3 to produce incredibly realistic videos with natively synced audio.

Similarly, OpenAI’s Deep Research agent has become incredibly helpful when I am writing a data-heavy article. While it might not always be 100% accurate for recent news, it easily saves me 10+ research hours every week.

If you want to leverage AI fully for your marketing workflows, I’d suggest making a habit of testing new AI features, models, and tools. I usually follow AI newsletters, such as The Rundown AI, to find these tools and stay up-to-date.

A Marketer’s AI Content Playbook

Given the rapid spread of AI-generated content, AI literacy is no longer a “nice to have” for marketers. Content marketers are expected to be AI-forward and integrate AI to amplify their marketing efforts.

However, there’s a right way to approach this problem. I’ve seen too many content marketers scramble to use AI to scale, but then spend hours fixing AI inaccuracies or editing content that doesn’t hit the mark.

My advice? Use AI prompts and tools to scale what’s proven in your content creation process, not replace what’s broken. If you use AI in an already inefficient system, chances are you’ll just make things more difficult for yourself.

Get your strategy right, then use AI to execute the mundane aspects of content creation.

Editor’s note: This post was originally published in May 2023 and has been updated for comprehensiveness.