AI is moving fast and life is getting busy

I normally try to get this newsletter out early Friday morning (like 6am) Australia time and yet here I am on Saturday morning writing this edition. I think I’ve managed to make my self imposed deadline about 99% of the time I’ve been doing the newsletter but work and life are starting to make that more difficult.

I don’t write this as an excuse. I hate excuses. I write it as an explanation for those who have their own routine that might be built around a quick skim of what is happening in AI and Travel at a certain part of your week.

My intention is to keep aiming for the deadline, but I need to prioritise Videreo’s customers first. The client list is growing, and I want and need to ensure that together we are putting out the best creator campaigns that will achieve their business goals for those clients. This has to be the priority.

Whilst it is true that the Videreo tech takes away 80% of the manual work for the brand, right now I’m personally sitting right on top of every campaign to understand how we can tweak and adjust to make that better and smoother and more efficient for everyone with each new campaign that launches. Ony by being in the trenches, can I know.

This week I got this message from a client: “Following the collabs we have created with you, we have seen some great engagement on social media post. Our followers has grown, we have had conversations in the DMs, received one booking and had a journalist reached out. Not everyone is a fan of content creators, but in this case by working with Videreo we have been able to ensure they share our values and integrity right from the outset. Thank you!”

This came from a client who wanted to fill a fully paid spot where the customer had cancelled last minute. They were paying their suppliers whether someone turned up or not. We filled the spot with a creator who was booked on 3 days after receiving the call for help. For the brand this was essentially free. That is just smart business. Their average booking value is $2K and they would not normally be an impulse purchase.

To turn these opportunities around, sometimes my focus needs to be there, with our client and delivery of this newsletter or podcasts might suffer. I hope that is Ok?

Videreo saves 80% of the manual workload associated with creator marketing, finds creators in destination so travel costs are avoided for brands and then measures everything against business goals.

Contact me to learn how we can make this happen for you.

This content is provided by the newsletter sponsor Videreo.com

Should AI be reactive or proactive

This post on LinkedIn this week caught my eye. I mean it is probably an ad - I didn’t really click through and investigate but I think the principle behind the message is correct.

The author Christian H suggests that currently most AI being shopped around is there to work reactively. Many businesses are looking to automate the existing process - rather than rethink the story in its entirety.

In this case the example is waiting for a customer to interact with an AI chat to express their problem in their hotel room. This mimics the old process. I have a problem. I pick up the phone and hope reception can remedy it.

“A guest checks in at 3pm. The room smells like the previous guest. The AC takes 40 minutes to cool down. The shower drain is slow.

They do not call the front desk. Most guests never do.

They leave a 3-star review on Thursday mentioning all three things. You read it on Friday.”

Instead, the author suggests that the AI sends a SMS to the client 15 mins after check-in to check everything is OK. This isn’t the old process because that just isn’t practical. Anyone who has checked in to a hotel knows the staff behind the desk need to spend 20 minutes clicking buttons on their mouse for each check in (why is that?) so there is no time for proactive follow up.

But AI is built for this type of follow up. It is conversational. It can do tasks. It can maybe even fix half the problems.

“It catches the problem before it becomes a review. It opens the door to an upsell before the guest has mentally checked out. And it tells the guest that somebody noticed they arrived.”

Being able to think beyond just accelerating an existing process is what makes us humans so valuable.

At Videreo this is our whole wedge. The old process of hiring creators is manual headhunting, one by one in a world of Instagram DM’s and ghosts. Just doing more of that, faster, isn’t really a solution to the fundamental problem. We flipped that into a “jobs board” where the brand puts out their offer - what they need and the compensation they are offering in return and a huge community always looking for work, assess the offer based on their own needs, location, brand fit and availability and then apply if it suits.

80% of the old workload magically disappears and satisfaction rises on both sides. Brands get who they need and fast and creators find work more easily.

Think first. Build second.

The true cost of building AI solutions

Alex Ragin this week posted about the real cost of implementing AI solutions.

“Many travel companies think AI-powered software will be cheaper.

That it will replace “expensive SaaS” and allow companies to build the same platforms for a fraction of the cost.

I think the opposite will happen.”

Alex’s thesis is that performing actions with AI and agents often creates multiple calls to AI API’s and when you hit scale - you might find a nasty bill you weren’t expecting at the other end. The modern version of bill shock.

“An “AI feature” is rarely a single request. It often involves multiple model calls, prompt chaining, retries, embeddings, vector searches, and sometimes several agents interacting with each other.

One user action can trigger 5 to 20 model calls.

Now multiply that by thousands of users.”

Right now AI is subsidised by those willing to put in the billions in order to extract the trillions later on.

It is certainly worth thinking about what happens when the subsidy runs out.

My own take here is that work and life are about to change in a fundamental way. Each individual will have the capability to spin up software to solve specific issues in front of them. The AI platforms will lock in every company on monthly enterprise contracts to extract their pound of flesh. The bigger companies will be subscribed to them all and be able to switch between platforms depending on what they are building and what the latest model capability is. The platforms will make plenty of money this way without restricting usage or making it beyond the means of most.

Microsoft has built a nice little business this way in the past. The fees will be higher than per seat Microsoft licenses - but they’ll also do a hell of a lot more.

The operational AI stack in travel

Skift’s Rafat Ali this week posted about how he took his regular scans of the jobs market from leading companies in travel to predict how companies are building out their operations with AI.

Rafat used Claude to help pull all the data together. I used GPT to just get the headline list here for you. Check the article for the full details:

• Organizational AI enablement

• Infrastructure

• AI governance & security

• Workflow deployment

• Truth & quality operations

• Commercial monitoring

• Industrial & project execution

• Service-channel transformation

• B2B platform leverage

The prisoners dilemma of AI investment in travel

A post that got a ton of engagement this week came from Felix Shpilman of Moses Capital about the “prisoners dilemma” currently happening at the top of travel.

“Every major player — Booking, Expedia, Sabre, Amadeus — would rationally prefer not to spend hundreds of millions on a product nobody is using yet. But if Booking invests and you don't, you lose the moment the market turns. That moment is impossible to predict, and you can't afford to miss it.”

The “product no-one is using yet” comment is about agentic commerce.

As Shpilman puts it: “$700M invested. 2% adoption.
Travel tech is caught in a prisoner's dilemma.”

Personally, I look back to the start of the internet for guidance here. I highly recommend the Acquired podcast series to dig in a little bit deeper here. The early Google story I think is really instructive. (Beware - these podcasts go for 4 hrs - no joke! Good for a flight.)

Google didn’t even start at the beginning of the internet. It started a few years later and seems to have done OK. Maybe even sitting back and waiting a bit was an advantage as other early movers flamed out.

Adoption is coming. Travel won’t be what drives it. Everyday uses will come first and train the masses to trust their AI. Companies like Visa will open the pathways, not Booking or Expedia. Travel will just get caught up in the wave.

The issues with early adoption are more around choosing a partner who might not make it (how many people are still using Netscape as their browser - for a while it was everyone…)

There are plenty of other opinions in the comments of the post for you to work out your own hypothesis.

Want travel advice? Ask Google Maps

Maya founder Joris Vanherp this week suggested the new AI super app in travel might be right under our noses.

Vanherp says: “Google isn't improving navigation. They're building an 𝐚𝐠𝐞𝐧𝐭𝐢𝐜 𝐥𝐚𝐲𝐞𝐫 on top of the world's richest location data. One that knows your preferences, anticipates your needs, and answers before you even think to call a travel agent.”

Joris uses a road trip from his home to the coast as an example. And I think probably Google Maps has always played a role in this type of journey. And they absolutely have always played a role for when people are in destination and navigating around unfamiliar streets.

He suggests this is a big problem for the itinerary planner type AI’s out there. I’m not sure about that because from my experience they play different roles.

Google Maps plays basically no role in me getting on plane to go somewhere. A small role in the hotel I choose, more if I have a specific neighbourhood in mind or site I want to be close to. And again little role in the bigger experiences I might be seeking when in a place.

It plays a massive role however in all my unplanned time and unplanned dining etc. What is near me now? Where can I get a snack? Where can I grab a drink before my next planned thing.

Not sure much changes here?

If you think someone (or everyone) you know or work with could grow from being more informed on the topic of ai + travel (or could use the training above) then please forward this email to them and they can click the button below:

Marketplace Spotlight: Travel Code

Travel Code CEO Egor Karpovich put out one of the most eye-opening posts I’ve seen in months a little while back.

Egor was one of the first I’d seen who had built and made operational a team of agents to help him run his business.

“We run Travel Code with AI agents. Not as a marketing talking point - as actual infrastructure.

Right now, a squad of AI agents handles our content creation, SEO monitoring, PR outreach, and competitive intelligence. They coordinate with each other, file reports, and flag anomalies without anyone prompting them.”

I needed to talk to Egor! the chat went live on the pod this week. More details below.

If you have a B2B business underpinned by AI and looking for people to notice you, you can sign up to the marketplace for peanuts (top right corner, 5 mins, bring your logo).

I’ve priced for bootstrapped startups but also accepting larger companies too.

Deal flow is collapsing in the age of AI

The always brilliant Lennart Dobravsky from TNMT this week put out a piece based on the research they had done of travel tech deal flow.

The headline: “Travel and Mobility Tech funding remains structurally weak.
2025 brought just $13.2B in funding and fewer than 400 deals (the lowest deal count in our dataset since 2016). So yes, the bearish narrative is real.”

AI investment however remains somewhat robust.

“AI is the major exception.
Travel is keeping pace with the broader market when it comes to AI startup funding. The capital is clearly flowing. The more interesting part: enterprise AI adoption across travel still remains low. That gap is where much of the opportunity lies.

→ Two AI themes stand out already.
First, autonomous driving. Second, customer service automation is becoming the clearest AI ROI case in travel, especially in online travel, where margins are tight, and efficiency matters more than ever.”

I question whether autonomous vehicle investment should be in travel at all. I think we will be artificially inflated if that is the case. Customer service is also not really travel specific either. It is industry agnostic and clearly the next layer after coding to be fully disrupted by AI.

I left my own thoughts in the comments - that maybe the lack of deal flow isn’t just coming from the VC side, but is also more and more a founder choice.

“Scratching below the surface a bit here on the low deal count - it might also be worth finding out how many founders these days are bootstrapping because the cost of building is close to zero.

Potentially it is the VC model that is collapsing?

In travel you need to prove traction just to get a look in on pre-seed where in other verticals you can get funded on team & vision.

In our AI in Travel slack group I’d say most of those that have built AI native and achieved traction have decided they don’t need the money outside a few B2C companies.

Some are taking in strategic partners to seed strap but these deals wouldn’t get picked up in normal deal flow research I wouldn’t think.”

I think for Videreo and others like us that have built AI natively from day one, it isn’t really cash injection alone that is interesting. We ship customer requests for product improvement the night they are requested and have them ready the next day. We don’t need money to build product which was usually where seed funding traditionally went.

Of course, like any business we’d like to be onboarding more customers faster and more people could help there, but we can fund that with cash flow.

For us to take investment at this point, I think it would need to come from people who had knowledge, skills and contacts that got us in front of more ICP’s. People who could instantly see the benefit of what we offer and where they are recommending Videreo to help their friends get an edge, not just grow the Videreo business. We’d love to be working with people who could immediately see their investment rapidly grow through their own influence, not just cash for cash sake.

If that is you - reach out and lets chat.

Slack Group!

Podcasts and Sponsors

Podcasts now on Spotify and Apple Podcasts:

New podcasts are now showing up on Spotify and Apple Podcasts for your easy listening pleasure!

This week I caught up with Egor Karpovich as mentioned above. This is bleeding edge stuff. One not to miss.

Most clicked last week was the link to Katalina Mayorga”s post about how she is currently building with AI and how you can follow along.

That’s it - you’ve made it to the end of this edition. If you’re thankful for this newsletter - you can always buy me a beer. Big thanks to Carl who bought 20 beers this week to say thanks. That means a hell of a lot.

I’ll be putting the result of the most clicked post in next week’s edition so you can see where others are focusing. If I’ve missed something, you’ve got a tip or any feedback at all - you can simply reply to this email and it will come straight to me. I’m doing this for You so please don’t be shy to tell me what you think