The Hidden Barriers to AI Adoption - And 4 Examples of How to Overcome Them

You and I live in a bubble, my friend. We feel like it is almost “AI or die” at this point, but here is the reality:

83% of companies prioritize AI in their strategy, but only 20-30% use it daily (source). That’s a massive gap. Why?

The gap between ambition and execution is massive, and it’s not due to a lack of technology—it’s about people.

To make this paradox more relatable, here are some of the arguments I have heard while ear-dropping to customer interviews, usability tests, and sales demos of my teams (with fine intentions to make their landing pages, sales decks, and sales scripts better):

• “Will this AI voice agent sound like a robot?”
• “I don’t understand how it is doing what it is doing, and that scares the s*** out of me.”
• “What if this AI SDRs spams my LinkedIn with generic messages and I get restricted from LinkedIn?”
• “This AI could literally replace 5 people in the existing process. How can we make them use this solution, if they know that one day could replace them?” - and my favorite:
• “Cool - I see this case study/pilot - but how can you guarantee this will work for me? Can we do the setup together?”

This is humans vs. AI IRL.

My job is to equip my teams to answer these objections that have little to nothing to do with technology per se.

We can solve these objections with some positioning and messaging tweaks, clever onboarding, marketing and sales tactics, and a lot of social proof.

In this post, we will dive into:

• 5 most common barriers to AI adoption defined by Ethical Intelligence Lab at Harvard Business School + how big corpos are dealing with them.
• 4 cases on how to bridge the AI adoption gap from small and medium businesses in my network that you can learn from.

If that sounds like a good use of your time and bandwidth,

Let’s rock and roll ??

This newsletter is sponsored by Brand24.

Interested in how to turn market chatter into a strategic advantage? Brand24 empowers you to tap into this potential with its social listening tools.

With real-time monitoring of over 25 million online sources—including social media, blogs, and podcasts—Brand24 keeps you one step ahead.

Whether you're tracking brand sentiment, evaluating campaign performance, or analyzing competitors, Brand24 offers actionable customer insights that fuel your strategic decisions.

Unlock an exclusive 10% discount on all Brand24 plans (valid for 90 days)!

5 barriers to AI adoption by Harvard Business School and enterprise examples of how to address them

Know your enemy. What are we up against?

In his extensive research, Julian De Freitas, Director of the Ethical Intelligence Lab at Harvard Business School, identified five key factors contributing to AI resistance:

1. Perceived Threat to Autonomy: To many, AI is a black box and people do not understand why it is doing what it is doing. Imagine if you are a pilot flying a plane … The more severe the decisions are, the pressing this problem becomes
2. Lack of Trust in AI: You know when your AI assistant hallucinates? There are (sometimes valid) concerns about AI's reliability, transparency, and potential biases.
3. Fear of Job Displacement: 74% of Americans are concerned that AI will cause job losses within the next 12 months (Forbes).
4. Ethical and Privacy Concerns: Fear of what happens to your data when you feed it to AI: worries about personal privacy or unethical use are another barrier to AI adoption.
5. Complexity and Understanding: If you do not understand what it can do (for you), you are unlikely to use this technology. Fastest growing European startup Lovable grew to $17M ARR in 3 months because the product is just wow - most of the AI products are very clunky and difficult to use.

Sounds familiar?

As technology matures, users are not just awed by anything it produces. We learned how to think critically about AI capabilities and its role in the evolution of our society.

Back in 2023, we were thinking, “Oh, cute, AI can draw a picture.”

But if I have to put $5000 budget in promoting that image for my Paris workshop tour, I will prompt the s*** out of it before I would be satisfied with the outcome.

Enterprises have developed many clever strategies and tactics to address these concerns:

• Most AI assistants like Siri, Alexa, etc. use female voices and even speech disruptors such as “hmmm, ammm…” to make them perceived to be more emotional and human.
• Netflix calls its recommendation engine machine learning, not AI.
• Google implemented transparent AI development practices, openly sharing algorithms and decision-making processes to build user trust.
• IBM developed the "AI Fairness 360" toolkit to detect and mitigate biases in AI systems, addressing ethical concerns and promoting fairness.
• Microsoft launched AI literacy programs to educate the public and demystify AI technologies, reducing fears associated with complexity.
• Salesforce introduced AI tools designed to augment human work rather than replace it, alleviating fears of job displacement among employees.
• Apple prioritized user privacy in AI applications, ensuring data protection and addressing concerns related to ethical and privacy issues.

Now, I know what you are thinking:

“Those companies have millions to do it - what can I do to bridge the AI adoption gap with my resources?”

No worries, I have you covered too.

Next, we will dive into more relatable examples for 90% of readers of the GTM Strategist.

4 cases of bridging an AI adoption gap from relatable companies in my network

Here is a curated list of companies I work with that might spark some new ideas into improving user experience, onboarding, and your positioning/messaging to bridge the gap from early adopters to slowly but surely reaching the early majority.

#1 Adapt the expectations and show how the output was generated

I was working with Linkbound this weekend, a tool to “turn likes into leads” on LinkedIn. You guessed it: a social selling tool. Their wow factor is that they filter leads for you based on your LinkedIn history and suggest the best social selling scripts based on detailed analysis of interactions.

Based on our past interactions (analysis of 174 reactions and 162 comments), AI “knows” Jonathan. I only provided a short CTA. Is it perfect yet? Nope, but it is perfect enough. I can do a couple of tweaks to make it “sound more like me,” and this one is good to go. I trust this v01 because it was suggested based on our past interactions and generated from a specific post he engaged with.

#2 AI does not have to sound like a robot ??

echowin is a platform for building AI agents (voice and chat). I will share two best practices from them - one internal and one external.

Meet Mars ??, one of my favorite coworkers. The team developed it using their platform as a Discord agent for internal communication. Mars makes v01 LinkedIn posts for the team, imports to-dos from our workshops to their ClickUp, saves resources we share to dedicated lists, and so much more. I love interacting with Mars. It makes our communication more effective and fun.

Now, to a more serious example: echowin offers a platform on which companies can build AI voice agents. One of their best use cases is Jiffy Lube, an American automotive oil change specialty shop chain. They have seen a 5X increase in customer appointments with echowin’s AI-powered phone agents.

But the number one concern when someone mentions “AI receptionist” is that it will sound like a dumb robot, especially in multi-ligual communication. To bridge this gap, the team has developed a “sandbox”, where you can have a real call with a voice agent you care developing. I was shocked when I tried to speak in Spanish with it and it spoke much better than I ??.

Show, do not tell!

Try echowin’s sandbox here: https://echo.win/auth/signup

#3 Make friends with humans - instead of threatening them

A Forbes survey found that 77% of workers worry AI will replace jobs within a year. It is natural to expect they will be reluctant to embrace the tools that put their livelihood in perceived jeopardy. But I am not on the “Matrix” team here.

While “replace humans with AIs to do grunt work” pitch actually works on some higher-up decision makers as a part of cost and error reductions mantra, it will backfire at the end-user level unless it is communicated as an enhancer, not a replacement.

The problem is that “those BDRs” are decision-makers in your buying committee.

This is the pitch that I like a lot.

Swan AI agent helps you to identify website traffic and start social and email selling sequences (warm outreach) to close more deals. Instead of “fire SDRs”, the team knows that whenever you get a response, a human will more likely than not have to close the sale. In this section of the website (positioning/messaging), they pitch that this technology is helping sales teams to be more successful, not replacing humans with AI bots.

Swan’s CEO and co-founder Amos Bar-Joseph generated $400k of qualified pipeline in 30 days all by himself using the Swan AI agent. His LinkedIn content brings a lot of traffic to their website, so Swan helps him focus on the conversations that matter and scale ultra targeted outreach.

“The smartest GTM teams are not looking to replace their SDRs with AI, they are looking to turn their sellers into 100x superhumans with human+AI collaboration.” - Amos Bar Joseph, CEO and co-founder, Swan

#4 Lead with use cases and templates, and fight the “blank page” problem ??

The reality is that the AI knowledge gap between the makers of AI and AI users is HUGE for most companies. Yes, people know that they should be embracing and learning about AI, but in reality, most decision-makers in business do not even use mainstream AI tools such as ChatGPT, Claude, and Microsoft Copilot in the moment.

Three practices that I have seen working well with my teams are:

1. White glove onboarding - meaning that “demo” is more of making a setup for them together - works great when you are starting out. Gets unscalable/too expensive later on for product-led companies with annual revenue below <$5000 per user, in my experience.
2. Upsell the education sessions and onboarding to learn how to do it together - I like this one because it is usually an extra revenue in most cases.
3. Provide them with templates, pre-setting options, and other “never stare at a blank page” solutions that you can think of. Do not make them think too hard. Drive the adoption by providing ideas on how users can benefit from your solution, educate them, and ultimately empower them to do a better job. In simple words, you tell me what to do with it so I can get what I want. ??

When we did a setup for Brand24 , a good onboarding was precisely the case. Look, I know very little about clipping and online reputation and social listening tools; all I wanted was to see how my brands are performing on other social media. I am very active on LinkedIn, but X, YouTube, Reddit and other places remained a mystery to me. If I cannot handle more than one channel properly, how hard can that be for bigger brands that have 40,000 mentions a day, not 40 like me? ??

What is genius here is that you can use Brand24 for competitor research instead of your brand if you are not big enough yet to care about this or are just up to getting some of their undeserved markets who are campaigning online, writing bad reviews, and desperately searching for alternatives.

Moreover, Brand24 also comes with an AI brand assistant and anomaly detector. See - I would not quickly figure that out as a noob user. That is good AI comm: what I can do with it (something that I actually care for), not merely what it does. I get bored if you are just namedropping features - on demos and in real life.

Very cool!

Hope you enjoyed our stroll down the AI adoption chasm. I will never forget what one of the founders I was working with said: "Technology is evolving much faster than social sciences. We have so little to operate there that it is best if we just share experience with each other." So, I would like to ask you to co-create this content with me.

Have you encountered resistance to AI in your organization?

What strategies have worked (or failed)?

I’d love to hear your thoughts in the comments.

Want more guidance on finding customers for your AI product?

I’m starting a live course on how to Launch and Scale AI Products Profitably at the end of this month. It will be a unique opportunity to work directly with me in a small group. Check out the whole curriculum and more on the link below, and use the code GOTOMARKET for a 100$ discount.

Learn more

If you have any questions regarding the enrollment, email me at [email protected].