Interviewing an AI version of HubSpot's founder was weird

Working at an agency sometimes doesn’t feel like real work - but interviewing an AI clone of Brian Halligan on our podcast didn’t just feel like not real work. It didn’t feel like real life.

I found myself nodding along, gesturing, reacting - like I was talking to a real human. Then, suddenly, it hit me: I was engaging in a conversation with what some would call a super-glorified Alexa. And yet, this wasn’t just any AI interaction - it was a demonstration of technology that has the potential to change the way we communicate, sell, and even exist in professional spaces.

It was weird. Very weird.

And it got me thinking… where does this technology go from here?

Wait, what's an AI Avatar, anyway? And who's Brian Halligan?

An AI Avatar is an digitally generated version of a real or fictional person that's trained to simulate voice, tone and expertise traits. Think the kind of thing you see in 'deep fake' videos crossed with a chatbot. But it's designed for business interactions rather than celebrities doing mad stuff.

The aim is to provide thoughtful, human-like interactions.

I'm sure there's a more correct AI definition but that's basically the basic principle.

Brian Halligan is the co-founder of HubSpot and a pioneer of inbound marketing. When he appeared on our podcast years ago, our agency’s co-founders had to fly to Boston to interview him.

This time? No flights, no time zones to juggle - just an eerily accurate AI recreation of Halligan, delivering insights on where inbound marketing is heading next.

But this wasn’t just a novelty experiment. It raised real, existential questions about trust, authenticity, and the role of AI in marketing.

Watch the interview by hitting play

Key insights from AI Brian Halligan on the future of inbound

When we originally interviewed the real Brian Halligan, inbound marketing was at its peak. But as AI Halligan pointed out, the landscape has changed. Here’s what his AI self had to say about the future:

1. Inbound is no longer enough on its own

   • AI Halligan noted that while inbound still matters, “content saturation and AI-generated search answers mean simply publishing blog posts and waiting for leads isn’t going to cut it.”

   • Marketers need to double down on community building, owned data, and personalized engagement to make their inbound efforts stand out.

2. AI will reshape how buyers research and make decisions

   • AI-driven search experiences, like Google’s Search Generative Experience (SGE) and ChatGPT-powered research, mean buyers can get comprehensive answers without ever visiting a brand’s website.

   • The challenge for marketers? “Optimising for AI’s consumption, not just traditional search engines,” AI Halligan suggested.

3. Trust will become the most valuable currency

   • Mistrust in AI-generated content and automation means that brands who showcase authenticity, expertise, and real human engagement will be the ones who win.

   • AI Halligan put it simply: “People trust people, not robots. Even as AI becomes more advanced, businesses need to double down on relationships, not just automation.”

The ethical dilemna: Can you trust an AI Avatar?

While our conversation with AI Halligan was fascinating, it also raised critical questions about trust in content.

If AI can convincingly mimic experts, what does that mean for credibility in marketing, sales, and business leadership?

• Does an AI avatar’s insights carry the same weight as the real person?

• Would you trust AI-generated advice over human expertise?

• If an AI version of a business leader says something, is it them saying it - or just an approximation?

These are questions that all businesses will have to grapple with as AI becomes more sophisticated.

Where could businesses use AI Avatars?

If AI Avatars can replicate knowledge, authority, and even personality, what’s stopping them from taking on bigger, more complex roles? Here are some potential - and mind-bending- ways AI avatars could shake up the way we work:

1. First-line customer support? Too basic. Try full sales conversations

• Chatbots already handle basic support, but what if AI avatars could handle entire sales calls? Imagine a world where your most experienced salespeople could clone their knowledge and pitch style to be available 24/7, across every time zone.

• AI Brian Halligan suggested that businesses will soon be able to “train AI reps to mirror their best human salespeople—scaling trust, not just automation.”

2. Attending meetings for you (and saying what you would say)

• Everyone has meetings that don’t need their full attention but require their input. Imagine an AI avatar trained on your knowledge, decision-making patterns, and communication style able to sit in, engage, and even negotiate on your behalf.

• Would you let an AI version of yourself handle internal status meetings? Client check-ins? Even leadership presentations?

3. Scaling thought leadership at a freakish pace

• Many business leaders struggle with time constraints for content creation. What if AI avatars could allow you to host webinars, deliver keynote speeches, or even engage in LinkedIn conversations - without you physically being there?

• At what point does this become an advantage, and when does it cross into deception? Would audiences really be okay engaging with thought leaders if they weren’t sure it was actually them?

Personally, I think it's way too early for some of these use cases. Here's how we'd advise deploying AI Avatars and human-like Agents for your company:

• Be honest.
  Always be clear that it's not a human. Honesty is vital.
• Gather extensive training materials.
  Make sure any Agents or Avatars representing you are well-trained to answer questions about your product, services and areas of expertise by training them on extensive knowledge bases and other materials.
• Set realistic expectations for yourself.
  Don't expect to use an Avatar like Halligan's clone any time soon for you or most of your workforce - he's got 15+ years of thought leadership content all over the internet attributed to his name. Most businesses will have to do a lot of 'offline' training of their own versions of this tool.

But, in regard to that last point, technology is moving so fast, I'm sure this will get solved sooner rather than later.

So, what's next?

AI avatars like the one we spoke with aren’t replacing human expertise... yet. But they will change the way businesses interact with customers, scale knowledge and build trust in an AI-driven world.

As inbound marketers, we need to think beyond traffic and leads. We need to ask: how do we use AI without losing trust?

Would you be comfortable talking to an AI version of your favourite industry leader? What about trusting an AI version of your colleague, boss, or business partner?

One thing’s for sure: this technology isn’t just changing inbound marketing—it’s redefining it entirely.

If you want to hear AI Brian Halligan’s full thoughts on inbound’s next evolution and form your own opinion on the future of AI avatars. Watch the full episode now...

Check out the full transcript to see what AI Brian Halligan had to say about the future of inbound and trust in the age of AI (yep, told you it was all a bit weird)

Brian Halligan (00:05.04)
It's Brian Halligan, how can I help?

Hi, Brian Halligan, it's Paul Mortimer from Avidly Talks. We're the five times Global Partner of Year of HubSpot. We're recording a podcast right now.

Brian Halligan (00:18.446)
Paul, great to hear from you. Avidly's been crushing it. Five-time global partner of the year with HubSpot. That's no small feat. I'm thrilled to be part of your podcast today. Let's dive in. What's on your mind?

You're one of the very few return guests to the show. You were on there originally when it was called Inbound After Hours and you talked about the future of inbound marketing. Do you remember that episode?

Brian Halligan (00:49.166)
Of course, Paul, I remember that episode back when it was called.

called Inbound After Hours, we had a great conversation about how inbound marketing was evolving. I think we touched on how the internet was creating this massive arbitrage opportunity for businesses to attract customers by being helpful and human rather than interruptive. It's wild to think how much has changed since then, but the core principles of inbound, creating value, building trust, and putting the customer at the center are still as relevant as ever.

What's been your biggest takeaway from watching inbound marketing evolve over the years?

Probably this trust thing that we'll dive into in this episode. But first, the same sort of topic as your original appearance or the real you appearing. The future of in-buy marketing now in 2025. What does that look like? Avatar Brian.

Brian Halligan (01:49.774)
The future of inbound marketing in 2025. It's all about trust and personalization at scale. Trust is at an all time low. People don't trust marketers. They don't trust sellers, but they do trust their peers, their communities, and the brands that feel authentic to them. So the game now is about creating real human connections in a digital world that's increasingly driven by AI and automation. Here's where it gets interesting.

AI is transforming how we approach inbound. It's not just about creating content anymore. It's about creating the right content for the right person at the right time. AI tools are helping businesses analyze data at a level we couldn't even dream of a decade ago, allowing for hyper personalized experiences that feel less like marketing and more like a conversation. But here's the catch. If you over automate and lose that human touch, you're toast.

People can sniff out inauthenticity a mile away, another big shift, the platforms themselves. Back in the day, Google and social media were open ecosystems. You'd create great content, optimize it, and let the algorithms do their thing. Now platforms are becoming more closed. They want to keep users on their sites, not send them off to yours. So inbound marketers need to rethink their strategies. How do you build trust and engagement when you don't control the platform? Community building is going to be huge here.

Brands that can create their own ecosystems, their own spaces where customers feel valued and connected are going to win. So inbound in 2025 is still about being helpful and human, but it's also about being smarter with your data, more intentional with your technology and laser focused on building trust. What's your take, Paul? How do you see trust playing into this new era of inbound?

Trust super key, obviously, I think that's something you touched on with sort of attributable pieces of content, being able to predict sort of returns from every bit of content you put out there on predictable search engines. That's all changed completely. But I wanna dive into something you said about building trust, being authentic, and people can sniff out what's inauthentic. I'm very, very, very hyper aware that I'm talking to an AI version of you now.

Paul Mortimer (04:09.107)
How does that compute with this area of winning by building trust?

Brian Halligan (04:17.752)
That's a great question, Paul. And I think it gets right to the heart of where technology and trust intersect. The way I see it, trust doesn't come from pretending to be something you're not. It comes from being upfront about what you are and delivering value in a way that feels real. So yeah, you're hyper aware that you're talking to an AI version of me, and that's okay. The key is that this conversation still needs to feel meaningful, helpful, and human. Think about it like this. AI, when used right, can actually enhance trust.

by making interactions more personal and relevant, but it's a tool, not a replacement for authenticity. If I, or the real me, were trying to pass this off as the real Brian without being transparent, that's where trust would break down. But by being clear about what this is and focusing on delivering value, we're leaning into the same principles that make inbound marketing work being helpful, human, and honest. It's the same challenge businesses face today. AI is everywhere, and it's tempting to over-automate

Brian Halligan (05:20.71)
are the ones that use AI to enhance the customer experience without losing that human touch. Like I've said before, people can sniff out inauthenticity a mile away. So whether it's a chat bot, an AI generated blog post, or heck, an AI version of me, question is, does it feel real? Does it add value? Does it respect the person on the other side? What's your take on that balance, Paul? How do you see businesses navigating the line between automation,

and authenticity.

I see a lot of AI garbage put out lazily and I see some good use cases of time being saved and efficiencies gained. One thing I'm really keen to explore and see what you or your still weird to call you, you, the avatar says here about I'm just conscious now actually I'm talking to you like you're a human, but I need to talk to the humans are the ones listening. So listen, what is a

AI sort of solution for attributing your marketing and content efforts in this age of walled social gardens, AI overviews on Google, in this new era of inbound marketing and content marketing. How do we attribute success to our content? Because trust is so hard to measure.

Brian Halligan (06:46.35)
Attribution in this new era is definitely trickier, Paul, with walled gardens like Facebook and Instagram keeping users inside their ecosystems, and AI-driven search tools like ChatGPT summarizing content without sending traffic back to your site. The old models of tracking clicks and conversions are breaking down. But here's the thing. While the tools and platforms have changed, the principles haven't. You've got to focus on what really matters, creating value and building trust. So how do you measure success when trust is intangible?

First, you've got to shift your mindset. Instead of obsessing over direct attribution for every single piece of content, think about influence. Look at metrics like engagement, brand sentiment, and community growth. Are people sharing your content? Are they talking about your brand in forums or on social media? These are the breadcrumbs of trust. Second, lean into first-party data. With platforms locking down third-party tracking, your own data, email subscribers, CRM insights, customer feedback, is more valuable than ever.

Use tools like HubSpot's CRM to connect the dots between your content and customer behavior. For example, if someone downloads an ebook, attends a webinar, and then becomes a customer six months later, you can start to see patterns in what's working. Third, experiment with new ways to measure impact. For instance, if you're running a campaign on a walled platform like Instagram, try using unique promo codes or trackable links to tie activity back to revenue.

Or if you're investing in AI generated content, A B tested against human created content to see which drives more engagement or conversions. Finally, don't underestimate the power of qualitative feedback. Talk to your customers, ask them how they found you, what content resonated with them and what made them trust you. It's not scalable, but it's invaluable. Trust is built one interaction at a time. And the more you understand those moments, the better you can replicate them. So.

Yeah, attribution is harder now, but it's not impossible. It just requires a shift in how you think about success. What's your take? Are you seeing any attribution methods that are working well for your team?

Paul Mortimer (08:55.128)
Those are good fallback principles that have been there all through the decades and decades and decades of advertising, marketing and consumerism as we know it. think that train of thought there about trust and how we measure that and brand sentiment and all these kinds of metrics are sort of new things to learn for a lot of new school marketers who were brought up in in-bank content era. Time for one more question, Avatar.

Brian and that is a tip, one tip for sales and marketing leaders to get their content engine right and set up correctly for this world in 2025.

Brian Halligan (09:39.98)
If I had to give just one tip, it'd be this. Focus on creating a content engine that's built for durability, not just virality. What I mean by that is think about content as an investment, not a campaign. The best content doesn't just drive traffic for a week or a month. It compounds over time like a good stock. For example, when we started HubSpot, we leaned heavily into blogging. Some of those early posts we wrote over 15 years ago are still driving traffic in leads today. That's the power of evergreen content.

But here's the twist for 2025. The formats and platforms have changed. It's not just about blogs anymore. It's about short form video, podcasts, and interactive content that meets your audience where they are, whether that's TikTok, YouTube, or even inside AI driven search tools like ChatGPT. So my tip is this, build a content engine that's flexible enough to adapt to new formats and platform, but grounded in the timeless principles of being helpful, human, and consistent. And don't forget to repurpose.

A single idea can become a blog post, a video, a podcast episode, and a social media snippet. That's how you scale without burning out your team. What's one thing you're already doing with your content engine that you think is future-proof?

Paul Mortimer (10:54.722)
The repurpose thing is the key takeaway there for me. This is where speaking to a lot of fellow like content marketers, there's a reluctance almost to fully embrace AI in process, but it's framing it to be a tool for you to produce that original thought leadership content and being a tool for you to repurpose what you've made into that scale. Hopefully this has worked this time. We've had some tech issues.

Um, I don't know. I'm going to give you the last word to Brian avatar. Brian end this episode, please. What are what are you going to close us with?

Brian Halligan (11:35.47)
All right, Paul, here's what I'll leave you and your listeners with. The world of marketing is shifting fast. yeah, AI is a big part of that. But at the end of the day, the winners are going to be the ones who stay grounded in the fundamentals, creating value, building trust, and connecting with people in a way that feels real. AI isn't here to replace marketers. It's here to amplify what we do best. Use it to save time, to scale your efforts.

and to repurpose like a pro, but don't let it strip away the humanity that makes your brand unique. And remember, the platforms, the tools, the algorithms, they'll all keep changing. What doesn't change is that people want to feel understood, valued, and connected. If you can deliver that, you'll thrive in 2020, five and beyond. Thanks for having me on, Paul. Always a pleasure.

Be sure to like, follow, subscribe and let me know what you thought of this. Let me know how weird this is for you and whether you found yourself like I did slipping into the habit of feeling like you're actually on a phone call with a human, not a computer generated avatar.