Prediction Not checkable as stated
Tan: Emerging browser-use AI agents will soon automate complex back-office workflows
“One thing that we're quite excited about on the horizon is this emerging browser use technology, whether it's, you know, some people call it computer use, some people call it operator, where AI agents will soon actually be able to work across a heterogeneous s…”
Insight
Tan: AI agents target labor budgets, making legacy software comparisons false
“So I think it's actually like a false comparison to look at the historical software incumbents and say, oh, this is The cap on what a company could become. I think there's just so much untapped opportunity that technology just wasn't able to penetrate before.”
Insight
Tan: AI excels at synthesizing unstructured data across disparate enterprise systems
“This is actually the type of work that AI is really good at handling. It is really good at taking very disparate amounts of information that is often unstructured in different formats across different systems. Synthesizing and structuring it, making sense of a…”
Assertion Not checkable as stated
Tan: Voice AI agents are now nearly indistinguishable from human operators
“Voice AI has been an incredible, I think, enabling innovation in the last few years where you can now actually talk to an AI agent on the other line. You may not even know that it's not a human because their conversational abilities and their intonation sounds…”
Assertion Not checkable as stated
Tan: High call-volume industries like logistics face the most AI disruption today
“I think where we're seeing the most disruption today Is industries that have very high call volume. We're seeing it a lot in logistics in particular, because if you think about how many different nodes are in a supply chain, there's so many people who have to …”
Insight
Tan: Building complex AI systems requires rare AI-native technical founders
“I think you have to be a really AI native technical founder to be able to understand how to leverage that, and that's actually just not a widely distributed skill set yet.”
Insight
Tan: The best AI startup opportunities target functions with clear ROI
“So we think that the best types of opportunities for people in that domain is Just really thinking about situations in which the ROI is so incredibly clear, which often means in types of work or types of functions where they have clear KPIs that you can assess…”
Prediction Not checkable as stated
Tan: Large BPO incumbents will face AI disruption only long-term
“This is a net new market doing very similar work to what they were offering, but to a different segment of the population. And so they might not see it in the short to medium term, but you know, if these companies do well, obviously they will grow up with thei…”
Prediction Not checkable as stated
Tan: AI coding agents will disrupt outsourced IT by empowering non-technical users
“We're seeing a lot of at a more horizontal level, like Coding agents just get a lot better and be able to empower people who maybe weren't as technical or maybe who weren't technical at all, be able to build full formed applications. And so I think that's actu…”
Assertion Not checkable as stated
Tan: Legacy RPA fails 20% of the time, requiring manual intervention
“So RPA is often very good for doing like 80% of the task, but then like 20% of the time that it fails, it's still a manual person who has to come in. So it's just not reliable enough to actually do the full task.”
Insight
Tan: AI startups should target a single, hyper-specific workflow first
“The way that we've seen it work best is when there's one very specific automation flow, at least to start, that a company can just nail. Meaning it's often industry specific. So you can integrate into all the core systems there. You can understand the context …”
Assertion Not checkable as stated
Tan: AI automation potential is an order of magnitude beyond RPA
“When you think about, like, the world of work that could be intelligently automated away, and the amount of time and savings Both employees and companies can get. It's just like an order of magnitude larger than what is currently possible.”
Prediction Not checkable as stated
Tan: AI will automate data entry and customer service in 10 years
“And if, let's say, 10 years from now, no one has to do manual data entry again, or no one has to, you know, get yelled at on the other side of the line for an angry, like, person in customer service, I think that'll be a win for everybody. Yeah. And then all t…”
Prediction Not checkable as stated
Tan: Video hardware-software models will succeed in heavy physical industries
“And we think that the same success could be found in other industries. Such as, for example, transportation, industrials, agriculture, mining.”
Insight
Tan: Flock Safety proved software sales models in difficult physical verticals
“I also think there's just like a business model understanding question, which a lot of people thought for a long time that selling into these Difficult verticals was tough. And I think now that we have companies like flock safety and proving that it is possibl…”
Assertion Not checkable as stated
Tan: Labor shortages force legacy physical industries to adopt software innovations
“I think there's a little bit of like a labor element of labor shortages in a lot of the industries that we were talking about that maybe opened the door to software innovation where people were not as used to buying software in the past.”
Prediction Held up
Tan: BPO market will grow from $300B today to $500B by 2030
“So the industry is valued at three hundred billion today with expectations to grow to over five hundred billion by 2030.”
Insight
Tan: High-value AI software flattens operational costs that traditionally scale linearly
“Another way to think about it is what sorts of operational work scales linearly as the company grows, meaning it's always going to be a consistent cost on the company. And, you know, as you get more customers, you have more customer support requests or As you …”
Assertion Not checkable as stated
Tan: Most intelligent automation starts with parsing messy unstructured data
“Almost Every intelligent automation path starts with some messy unstructured data that you need to pull key outputs from.”
