Aboard’s AI-powered bookmarking and project app is a new spin on a chatbot

Aboard is not an easy app to explain. It used to be simpler: At first, it was a way to collect and organize information: Trello meets Pinterest and that spreadsheet full of links you use to plan your vacation. The company’s founders, Paul Ford and Rich Ziade, are two longtime web developers and app creators (and, in Ford’s case, also an influential web writer) who previously ran a highly regarded agency called Postlight. They’ve done some interesting work on analyzing websites to extract useful information, and they’ve built a handy visual tool to display it all. “People love saving links,” Ziade says, “and we love making those links beautiful when they arrive.” Simple!

But now I’m sitting here in a shared workspace in New York, minutes after an earthquake and days before Aboard’s biggest launch yet, and Ziade is showing me something very different. He opens a beta version of the app and clicks a button, and after a second the page begins to change. A database appears out of nowhere, with a number of categories (year, title, genre, etc.) that start to be populated with a number of well-known movie titles. The painting, as Aboard calls it, is titled “Movie Night.” With one click, Ziade has just created – and fulfilled – a way to track your viewing habits.

Perhaps the best way to explain the new Aboard is not as a competitor to Pinterest but as a radical overhaul of ChatGPT. When Ziade created this map, all he was doing was querying OpenAI’s GPT-3.5. The company’s chatbot could have returned some of the same movies, but it would have done so with a series of paragraphs and bullet points. Aboard has created a more engaging, more visual AI app – and made it so you can turn that app into anything you want.

Ziade and Ford imagine three main things you could do with Aboard. The first, “Organize,” is closest to the original vision: Ask the tool for a bunch of things to do in Montreal this summer, and it will populate a table with popular attractions and restaurants. Ask Aboard to meal plan your week, and it will create a board segmented by day and meal with well-formatted recipes. The second, “Search,” is similar but a little more exploratory: ask Aboard to retrieve the most interesting links on African bird species, and it will put them all together for you to browse at your leisure.

Like any AI product these days, this one is sometimes cooler in theory than in reality. When I ask Ziade to create a chart of the important technological moments of 2004, he puts a number of them together into separate cards: the Google IPO, the launch of Gmail, the launch of the iPod Mini. And then the iPod Mini launches again, then another time, then three more times. Ziade and Ford both laugh and say that’s what they see all the time. A few times a demo fails, and every time Ford says something like “Yeah, that happens when you ping the models.” » But he says it’s also improving quickly.

The third use case, which Aboard calls “Workflow,” is where Aboard believes its real business lies. Ziade does another demonstration: he enters a prompt into Aboard, asking him to set up a claims tracking system for an insurance company. After a few seconds, it has a fairly simple but useful table for processing claims, as well as plenty of sample cards to show how it works. Will this be perfect and powerful enough for an insurance company to start using as is? No, but it’s a start. Ford tells me that Aboard’s job is to build something good enough but not good enough either – if the app can work just well enough to let you customize it all the way to meet your needs, it is the objective.

This is ultimately a very commercial use case and puts Aboard in loose competition with the Airtables and Salesforces of the world. Ziade and Ford are upfront about it. “We want to be in a professional environment,” says Ford, “that’s a real goal we’re aiming for. This doesn’t necessarily have to be for large companies, but definitely for small teams, nonprofits, things like that. He believes Aboard can sell to companies by saving them a lot of time and having meetings to figure out how to organize the data and launch it immediately. An Aboard can be a table, a list, a gallery, etc. it’s a flexible enough tool to handle most types of data.

I don’t have any particular professional use for Aboard, but I’ve been testing the application for a while and it’s a very smart way to rethink the output of a large language model. Particularly when combined with Aboard’s ability to analyze URLs, it can quickly gather some really useful information. I’ve been saving links for months as I plan a vacation and asked Aboard to build me a project planner to handle a large bathroom renovation. (This is all very exciting.)

Just before Aboard’s AI launched, I tried creating another board: I asked the AI ​​to create “a board of Academy Award-winning films, with stacks for each movie genre and tags for Rotten Tomatoes scores,” and Aboard got to work. He came back with stacks (Aboard’s language for sublists) for six different movie genres, tags for different score ranges, as well as running times, posters, and Rotten Tomatoes links for each movie. Were all the selected films winners for Best Picture? No! Have the grades been as good as ever? No! But it still seemed like a good start – and Aboard still gives you the option to delete the sample maps it generates and start from scratch.

Aboard is just one of a new class of AI companies, ones that won’t try to build another big language model, but will instead try to create new things to do with those models and new ways to interact with them. Aboard’s founders say they eventually envision plugging into many models, as those models become, in some cases, more specialized and, in others, more commoditized. In Aboard’s case, they want to use AI not as a responder but as a software generator. “We still want you to go to the web,” Ford says. “We want to guide you a little bit and maybe give you a helping hand, but we’re software people – and we think the ability to get started really quickly is really, really nice.” Aboard’s founders want AI to do the work, so you can just get to work.