Five examples of mastering algorithms
We live in a world where algorithms quietly shape our lives in ways we might not even recognize.
They feed us ads that influence what we purchase and where we live. They shape what we value and our taste in everything from food to music to humor. And, in some cases, they can bring together complete strangers who would have never met. Today, 60% of couples rely on algorithms to meet their spouse (via online dating apps).
In the balance of power between individuals and tech companies, big tech today has far greater influence. But individuals are still finding ways to harness algorithms for their own gain.
In this newsletter, we explore five examples of how individuals are asserting their agency over algorithms or opening themselves up to algorithm-created possibilities and connections. Algorithms might be powerful, but that doesn’t mean users are powerless.
// #1: Algorithms & Unexpected Connections
It’s easy to forget the original promise of social media: to connect people. Amidst all the ads and data harvesting, people are using algorithms to open themselves up to opportunities and increase the likelihood of chance encounters—from finding love via online dating sites to networking for their career. In the US, 30% of adults, and over half of adults under age 30, use dating apps.
- Happn, a dating site, uses a location-based algorithm to match users based on where their paths have crossed. It has over 150 million users globally and has led to millions of dates. Its power is in helping increase the chances that someone can find a way to connect with the people they might bump into as they go about their day.
- A 2022 study by researchers at MIT found that algorithms on LinkedIn that suggested people who were loose acquaintances led to greater job mobility than algorithms that just promoted strong ties and close connections. The conclusion: algorithms that served up unexpected connections could lead to greater career advancement and more opportunity. Of note: the A/B test central to this research spurred questions of ethics.
// #2: Algorithms & Side Hustles
There are 41 million gig workers in the US. In many cases, their employment and compensation are controlled by algorithms. In the book Inside the Invisible Cage: How Algorithms Control Workers, author Hatim Rahman argues that the algorithms that underpin the operations of platforms like Uber, Lyft, and DoorDash create an “invisible cage” where workers are controlled by opaque systems they neither understand nor influence.
However, a New York Times article last month profiled how some gig workers in New York City are making up to $6,000 per month by gaming the algorithms from Citi Bikes, a bike-sharing program. Citi Bikes compensates gig workers to move bikes from full stations to empty stations to ensure there are bikes at every station. Soon, people noticed an opportunity: the algorithm that dictated compensation worked on a sliding scale. Under the right conditions, gig workers could get paid 3-4x their normal rate for repositioning a bike. Working as a team, they quickly shuffled bikes to and from stations just blocks away and made $4.80 for each one-way trip. For one off-Broadway actor in New York, it was enough to pay rent and cover bills.
Brent Mittelstadt, a philosopher at the University of Oxford who studies the ethics of algorithms, concluded that this “scam” was harmless. More importantly, he argued, any time gig workers can reclaim some power from algorithms, it’s a win.
//
“The most common way people give up their power is by thinking they don’t have any.”
//
// #3: Algorithms & Break-out Stars
Algorithms can create kings and queens out of mere mortals. The internet is full of people who became famous overnight due to one song or video going viral.
- Lil Nas X—now famous for his hit song “Old Town Road”—worked the TikTok algorithm aggressively to promote this first song. “I promoted the song as a meme for months until it caught on to TikTok and it became way bigger.” he said. That effort paid off, and the song soon went viral. Lil Nas X rode his algorithm-driven success to the top of the Billboard charts (and two Grammys).
- Sheehan Quirke went from working at McDonald’s with zero followers to over 1.5 million X followers in 18 months. Quirke, known as the Cultural Tutor on X, explores topics like historical figures, classical music, architecture, and art history. Like Lil Nas X, his road to stardom started inauspiciously. His first tweets were met with relative silence. But then the algorithm took one tweet viral, and then another, and soon, his followers grew exponentially.
Lil Nas X and Quirke weren’t reclaiming their power from algorithms that had stolen it. Instead, they studied how the algorithms worked and then used them to catapult into greater fame and opportunity.
// #4: Algorithms & Data Boundaries
For some in the artist and author community, the fact that AI algorithms are being trained on their original work without permission amounts to theft. A new tool, called Nightshade, enables artists to add imperceptible changes to their digital art. Those changes can mess up training data and be challenging to remove.
Another tool, Glaze, protects artists by masking their unique style. Similar to Nightshade, it adds slight changes to a piece of art so that the AI algorithm interprets the style in a completely different way (ex. an image that is realistic is interpreted by the algorithm as abstract).
Such small acts of subterfuge might pale in comparison to larger efforts to file copyright lawsuits against tech companies like OpenAI and Microsoft, like the lawsuits from major US newspapers earlier this year claiming copyright infringement. But they’re giving artists a newfound power to push back against tech giants and reclaim some control over their intellectual property.
// #5: Algorithms & Information Sources
For the everyday internet user, algorithms have the most influence over what content shows up on social media and news feeds. Thanks to algorithms, the content we consume online has become increasingly personalized for us. But some users are fighting back:
- They’re changing browsers to ones like DuckDuckGo and Brave that emphasize privacy and ad-blockers.
- They’re changing search engines and prioritizing ones like Freespoke that provide more balanced perspectives.
- They’re “ruining their search history” by using a tool that generates random search queries to prevent Google from building an accurate picture of who they are.
An article in WIRED earlier this month outlined various ways that users can stop their data from being used by AI, from changing settings on their LinkedIn profiles to submitting a request to Slack to ensure their private messages aren't scraped.
// A people powered internet
Project Liberty Founder Frank McCourt wrote in Fast Company earlier this year that “our current reality is…best described as digital feudalism. Like poor, powerless subjects of monarchs and aristocrats, we are serfs, subjugated by a small group of companies that have exploited a feudal internet architecture.”
Tech companies wield outsized power—which is why the efforts of individuals who attempt to reclaim control of their digital experience are no substitute for bigger change: regulation, policy change, and by-design changes to platforms and algorithms that prioritize interoperability, user privacy, and copyright enforcement.
The People’s Bid to acquire TikTok is one such structural change to rebalance the power between tech companies and individuals.
There are hundreds of organizations, like those in Project Liberty’s Alliance, who are committed to building a better web, but that effort won’t happen overnight. In the meantime, individuals are not without agency to harness the potential of algorithms for their benefit.
As the poet and author, Alice Walker, once said, “The most common way people give up their power is by thinking they don’t have any.”
