The overseas niche AI track "raising 100 million USD", how does Laurel analyze temporal causality?

Laurel is building the world's first AI time platform to address the causal relationship between industries that can't accurately link time investment to business outcomes. This article is from Leo, an article written by Deep Thinking Circle, collated, compiled and written by techflow. (Synopsis: When the world rushes into AI, why is Apple still standing still?) Laurel is using AI to solve a trillion-dollar industry pain point: making knowledge workers' time visible, measurable, and optimized. Have you ever wondered why the manufacturing industry can calculate the cost of producing a car with great precision, and the retail industry can accurately track the inventory of each item, but law firms, accounting firms, and consulting firms are ignorant of their most important resource: human time? This question bothered me for a long time until I learned about Laurel's $100 million Series C funding round. The company is using AI to solve a trillion-dollar industry pain point: making knowledge workers' time visible, measurable, and optimizeable. I dug deeper and found that Laurel wasn't just doing something as simple as time tracking. They're building the world's first AI time platform to try to solve what founder Ryan Alshak calls the "time intelligence challenge" — the inability of knowledge-based industries to accurately link time investment to business outcomes. In the age of AI, quantifying and understanding human capital has gone from being the icing on the cake to being a "life and death" business need. The round was led by IVP, with participation from GV (Google Ventures) and 01A, and new investors also included celebrities such as DST Global, OpenAI's Kevin Weil, Alexis Ohanian, GitHub CTO Vladimir Fedorov and others. The Pain and Awakening of Six-Minute Bookkeeping The root of the problem can be traced back to the way the professional services industry has worked for decades. Lawyers, accountants, and consultants need to record their working hours in six-minute increments so that clients can pay by the hour. Ryan Alshak experiences this pain acutely as a lawyer: "It's like on a busy Saturday night when I'm a chef cooking for 500 customers, but at the same time asking me to keep track of every ingredient I use, which is both distracting and dehumanizing." I can understand the frustration. Imagine you've just completed a complex legal analysis and your thoughts are at their clearest, but then you have to stop and remember: How long did I just spend looking at the information? How many minutes did it take to write this memo? What was discussed on the call with the customer? This forced interruption not only affects efficiency, but also makes professionals feel like factory workers being monitored rather than experts providing intellectual services. Alshak's epiphany moment came simple: "Why should I tell the machine what I did at work instead of letting the machine remind me what I did?" Behind this seemingly simple question lies a counterintuitive insight: lawyers, accountants, and consultants actually have an underbilling problem because they forget a lot of the work that has already been done. If you can make more profits for the buyer (business) and save time for the user (professional), this is the perfect foundation for building a company. This pain point is far more common than I thought. According to Laurel, the average professional recovers more than 28 minutes of billable time per day, which were previously lost due to missing records. At an average hourly rate of $375, that means each professional generates an additional $175 per day for the company. For a large firm with hundreds of professionals, this number is quite staggering. Four keys to AI redefining time tracking Laurel's solution sounds intuitive, but it's an extremely complex technical challenge to actually build. I've learned that to truly automate end-to-end schedules, there are four key technical issues that need to be addressed, each of which has a fairly high technical threshold. The first challenge is digital footprint tracking. Laurel must be able to integrate with every digital program used by users, including Slack, Microsoft Outlook, Zoom, and other work tools. Only when AI can "see" all the work activities of professionals across platforms can it accurately reconstruct their work trajectories. It's like installing a ubiquitous but completely insensitive surveillance system in the user's digital work environment, capable of recording every click, every document edit, and every phone call. The second level is deep integration of AI applications. Laurel uses a variety of AI techniques to process these digital footprints: data clustering algorithms categorize related work, machine learning models assign work to relevant customers and projects, generative AI creates job descriptions, and finally encodes and classifies work through machine learning. Instead of simply applying a ChatGPT interface, we build an AI system optimized for professional services workflows. The third link is the delicate balance of human-robot collaboration. A draft calendar is generated for users who can add, delete, or edit content. This "human-in-the-loop" design both guarantees accuracy and allows AI to continuously learn and improve. Every interaction of the user makes the system smarter, which creates a positive cycle. The fourth step is seamless integration with existing billing systems. Once the user confirms the schedule, the system will automatically push the data to the firm's billing system, leaving the back-office management unchanged. In this way, the work experience of professionals has changed from "filling out the schedule" to "auditing the schedule", which greatly reduces the psychological burden. The ingenuity of the whole process is that it does not force the user to change their work habits, but works silently in the background, and in the end only requires final confirmation from the user. This design philosophy embodies deep product thinking: the best technology should be invisible, it should make complex things simple, not add a new learning burden to the user. From legal tech loser to pioneer in the AI era, Laurel's success has not been easy, in fact, it has undergone a complete rebirth. The company was originally founded in 2016 under the name "Time by Ping," but struggled in its early years. Alshak candidly acknowledges two main problems: the over-focus on the legal single market, and the lack of maturity of natural language processing technology at the time. The turning point came in 2022, when Alshak gained early access to OpenAI GPT-3, he made a bold decision: suspend all work and completely refactor the product. This is an extremely rare move in startup circles, and most people will tell you "never rebuild, keep iterating". But Alshak has chosen a path that goes against conventional wisdom, which I think exemplifies true entrepreneurial spirit – a willingness to take big risks for a bigger vision. When ChatGPT was launched in November 2022, the perception of AI in the entire market turned upside down...