Super interesting post, Elizabeth! Of all the airlines, I can see Southwest leveraging AI the most effectively, simply because of their focus on lower prices and increased value for both shareholders and customers. The customer segmentation piece is particularly interesting to me because I used to carry out segmentations for all my clients when I was a consultant. To be honest, I’m still a bit skeptical that an AI could carry out the segmentation as well as trained human could, but I can absolutely see the value in having an AI run the algorithm that an operator could then tweak and observe — while most of the value from segmentations comes from a human’s interpretation of what the data means, getting a program to run the time-intensive portions would be a massive boon to companies.
Super interesting post! As a former consultant, this model resonates strongly with me, as which project you get staffed on has massive impact on how long you stay in consulting for. I wonder how this might impact consultant happiness over time: while I do agree that engagement in a project is a major driver of satisfaction, I also know many of my colleagues would relish their time “on the beach”, particularly when they just got off a tougher / longer engagement. With the increased staffing efficiency of this model, there would naturally be less down time in-between projects — I wonder how that might drive tenure in the long-run?
Super interesting post, Isabella! The fact that AI can now generate music is awesome, if not wholly unexpected. A few years back, there was an MRI study done by psychologists that showed how our brains light up to different types of music, and how that data could be fed into an algorithm to select already existing songs to queue up next. I think the use of AI is particularly clever by Spotify because, as you said in your post, it helps them avoid the massive royalties they pay the artists (something that has kept Spotify in the red until recent years).
I’m actually pretty shocked at how well the AI was able to capture exactly what you were going for in this post: I guess the more words you use and the more context you can feed the machine, the more accurate the results will end up being, which makes sense. I wonder what the future applications of this are, as the technology becomes more sophisticated?
I love the use of stock imagery by Craiyon! It looks like so many images I used to have to upload into PPT decks when I was first starting as a consultant; also, the fact that so many of them seem to have people vaguely pointing at random, business-like objects (e.g., emails) is a nice throwback 🙂
This is actually horrifying, and maybe doesn’t bode well that the AI attributes such monsters to HBS’s mantra! I think you raise a really interesting point about the assumed gender the AI is applying — it likely is, given it’s searching through years and years of leaders who have graduated from HBS, the vast majority of which are men. Really interesting to consider the implication it has on HBS’s history of admittances.
Super interesting, Isha! I love Duolingo, mostly because of how it takes simple concepts and breaks them down into bite-sized lesson “chunks”. I read recently that Duolingo has just released its first Math product, taking a similar approach to what it’s done for languages — given math presents a different set of opportunities (e.g., math doesn’t require expensive translation like language does) and challenges (e.g., math is likely a less sticky product because it’s arguably harder to make more interesting), how do you think that fits into Duo’s current business model?
Awesome article, Isabella! I know you mention Rappi’s first mover advantage inured it from competition from Uber Eats; do you think this will continue to protect it over time? I find functional services like Rappi’s interesting, because they seem so prone to disruption or a price war, I was curious if there was any sort of complementary service Rappi offered that gave it a more unique value proposition?
Super interesting article! I’ve never used Fiverr, but I’ve always wondered about how it defends against disintermediation (i.e., the “switching costs” you outline above). It could anonymize users and gig workers, ensuring they could only exchange messages via the platform, but that friction would likely push users elsewhere. Do you think there are any complementary services it could offer that would make the platform stickier?
I really enjoyed reading the blog post, LMB — thanks so much for sharing. I think your/Brian Chesky’s perspective on Airbnb sitting at the intersection of art and science is spot-on: the landing page is beautifully laid out, search options are intuitive and human-centric, and the information is usually presented with the most important facts first (e.g., is the host a super-host, are the dates flexible). I think “using the customer’s data to create the perfect customer experience” is exactly what data should be, and is the right framing for any company trying to leverage data. It reminds me of the Marriott case we discussed last year in RC, and how Airbnb had gained so much market share partly because it was able to provide perfectly curated vacation packages to traveling millennials. Thanks again for the interesting read!
Super interesting blog post, Gigi. In particular, I really liked how you tied in our classroom understanding of building the AI assembly line and showing how Zhuiyi is applying it in real-time (e.g., how Zhuiyi has trained its AI to read tens of thousands of written conversations with call centers). I always wonder if AI can ever truly replace call center associates; anecdotally, whenever I call my bank or credit card company, I immediately ask the AI to send me to a human representative, because I’m skeptical of how adept the AI is at parsing my requests. Either way, I agree that it represents a tremendous opportunity for disruption in the market!
Super interesting post, Patric. I’m particularly interested by the last piece, on data security and privacy, because I wonder if the precedent Flo has set with Facebook might leave the door open to further litigation. For example, I’m wondering if this data could ever be misused to discriminate against hiring women trying to get pregnant (e.g., if the data ever got scraped or leaked, and somehow future employers were able to use it in the wrong way). I know similar claims have been brought against 23andMe with their repository of genetic material, and so I wonder if Flo might face the same issues? Either way, great blog post — I really enjoyed the read!