Algostorm: Algorithm Storm Social Manipulation, Trendjacking Botnets & Government Countermeasures - E454

​​Jeremy Au discussed the increasing intensity of "algostorms" - unpredictable social media “weather systems” driven by algorithms that prioritize engagement over truth. The Internet’s transition from human-centric, chronological forums to a volatile online environment where individuals are caught in fluctuating must-discuss trends manipulated by botnets, corporations, and professional influencers. He also drew parallels to the stock market, where algorithm-driven trading can cause flash crashes, leading to rapid market drops and recoveries. Some people trendjack to push their agendas, others unconsciously follow what seems to be majority opinion, and some seek refuge in genuine human interactions. He pointed out government actions to regulate content platforms and deploy Internet “circuit-breaker” blackouts, across Singapore, India, Myanmar, Indonesia, Iran and Bangladesh. Acknowledging the power of online influence is the first step to taking back control over one’s reactions and gaining intentionality in the digital sphere.

Please forward this insight or invite friends at https://whatsapp.com/channel/0029VakR55X6BIElUEvkN02e

Supported by Evo Commerce!

Evo Commerce sells premium affordable supplements and personal care electronics, operating in Singapore, Malaysia and Hong Kong. Stryv brand sells salon-grade quality products for home use and using direct-to-consumer channels through its online retail channels and physical shops. bback is the leader in hangover remedies in over 2,000 retail outlets across the region. Learn more at bback.co and stryv.co 

(01:09) Jeremy Au:

So I'm using my phone, I'm scrolling. I'm reading. I'm smiling. I'm laughing. And suddenly I find myself reacting. I feel angry, I feel sad, I feel concerned and then, I keep reading more about that thing and then I get deeper and deeper into it and time flies and I realize it's like an hour, two hours later and I'm still in bed reading this thing.

This thing is a news article or a kind of current affair or some topic that I'm thinking about. And then I find myself a few days later wondering to myself, what happened? That was something I cared about a few days ago, but now, I don't care about it at all. It wasn't impactful to my life. It was something that I read, but it wasn't something that truly mattered. I find myself in this cycle over and over again when I use the internet and I got myself wondering what exactly is happening here.

An aspect that I think about is what I call the "algostorm". An algostorm is the macro chaotic and uncontrollable environment created by social media algorithms that prioritize engagement, as well as other bots, influencers that are interacting with one another in complex, weather-like conditions. The storm is characterized by the rapid spread of trends, ideas, or emotions. The crux of it is that individuals have much less agency and power against this huge flow of information or opinions that seem to represent the mass society.

Public sentiment can rapidly change. Viral topics can quickly rise and fall. Individuals are caught in a whirlwind of changing narratives, resulting in a volatile and unpredictable online atmosphere. This is similar to how I might feel staying at home. Inside, at home, it's calm, it's protected, it's a safe harbor. And then I look out the window and I see it's a giant storm going out. And I know that the weather is having rainy, stormy, dark. And so I have a choice whether I want to walk out the house, into the storm, and so I better wear my umbrella, my raincoat, my boots, so that I don't get wet, I can walk around in the rain, or I choose to stay home.

The crux of it is that I don't look outside and say, I can change the weather, because I can't change the weather. The storm may be around for one hour, one whole day, several days, there's a monsoon or hurricane, but I can't change the weather. I have a choice about how I react to the weather. The weather is not controlled by me, but it's controlled by a whole web of complex factors. That's how I think about algorithm storms.

So why is this happening?

Historically, the internet used to be a pure human place where only humans were on the internet. Everybody was there because they paid for the dial up and they dialed into the forums and the forums used to be text based only and they were all in a chronological form or reverse chronological. So internet forums were effectively discussion threads and people were rewarded for following the discussion because if you didn't follow the discussion thread, then people would just ignore you. So people could read those internet forum discussions in a linear, chronological way that reflects the history of how you and I perhaps would have a relationship, which is that if I was to have a conversation with my friend today, it would be the last known conversation out of a series of 10 or 20 or 30 conversations over the past six years.

However, today's social media is really driven by engagement-focused algorithms. What that means is that if you and I had a set of linear conversations on the internet, for example, the algorithm would chop up and be focused on taking the most interesting snippets of our conversation and presenting them based on different audiences. So an audience person who is listening to it may say, "Oh, I'm interested in cooking." And therefore, when Jeremy discussed cooking with his friend seven years ago, that post is now elevated and escalated over Facebook or Instagram or TikTok. That's obviously beneficial for a creator because out of the world, today, there are millions of people and there's no way for me to personalize my conversation with anybody, to be tailored at 1,000 different personas in the world. However, on the receiving end, when you are receiving in a platform that is giving you a spread of engagement, then their job is to really push a lot of different buttons and make sure that every post pushes a button within you. And if that button doesn't get pushed in you, then they know that because you're not paying attention, you're not engaging it, you're not searching more for that information. And so they will look for a different piece of information that will drive a different feeling or emotion or issue because they're in search of that next button. And so these algorithms have taken over all timeline-based feeds. There was a big hoo-ha years ago when Facebook kind of said, "Hey, we no longer do chronological or reverse chronological feeds. We're now focused on algorithmic feeds." We see that with Instagram where it used to be about the posts within your social network.

And now, of course, they're adding more ads that come from advertisers, but what's interesting is they decided injecting a lot of posts from people that you've never heard of. Influencers and other digital creators and people in your very far networks, friend of a friend of a friend. Because they feel that, your old social network that was generating content wasn't doing it in a way that was spicy enough to keep you engaged to the platform. This algorithmic change is understandable from a social media perspective because they are trying to have you spend more time on a platform instead of spending time on someone else's platform, or spending time in real life, or spending time sleeping. I mean, I always remember that Netflix, the CEO said, "Hey, our biggest competitor isn't other TV networks. The biggest competitor is sleep." And that's kind of hilarious, if you think about it, and really dark because what they're basically saying is our job is to rob you of your health that involves sleep and put it into spending more time with us watching TV and movies that have been driven to you because we want you to spend as much time as possible watching TV at the expense of everything else. So it's not as if you have a budget of, say, one hour of TV per day, and then you eat that one hour into Netflix from a normal TV to Netflix. It's now Netflix versus everything else in your life, whether it's cooking, sleeping, or hanging out with friends. So the algorithms are really prioritizing engagement and in the war with everything else in your life.

(07:10) Jeremy Au:

And as a result, we have created this set of incentives for companies and professionalized influencers to share the content, myself included, as somebody who is sharing my point of view, but, that has created a new norm where, you're not blogging, but you're just creating content. And so for myself, obviously, I'm sharing long form content and I don't really do very well in terms of engagement. This is for people who are interested in hearing this linear chronological view? Is this that, if you look at the latest viral trends, you see people sponsoring ads in sport shoes for viral short moments, these fads. There's a whole corporate, capitalist monetary system that's really driving a lot of people to inject new content that is designed to be very engaging. And this creates in the complex pieces where suddenly, there's all these crazy memes that used to exist, but now they're getting super viral and the people making these memes, many of them of course are amateurs, but many of them are also professional folks who are sitting down and saying\, how do I make this a meme? How do I make this very clear? And we see this most obvious, for example, the political arena, because there are millions of dollars that are at work making professional posts, posting on Reddit, posting on Instagram engaging influencers, and obviously engaging a lot of community folks who are doing this out of their own free goodwill, but there's a big part of it that's actually being politically generated.

And we see this even in a case where in the Indian elections that now deep fake voices are now endorsed and used by candidates. So these are official candidates who are officially paying for AI to duplicate their voice because they want to send a personalized voice message to every individual person, voter, persona, a different language, a different area, neighborhood, they want to personalize it. And so, another aspect is that there's also botnets where it's not just humans were doing it and not just professional humans were being paid to do it, but there are also bots that are being paid to champion, to engage, because they're trying to get the algorithms to prioritize the content that they're focusing on.

And we see this in the Russian botnets. We see this in the American botnets. We see this in the Chinese botnets. But everybody is trying to push for certain agendas because they know that these algorithms are gameable. And the social media companies that have these algorithms know that they're being gamed, and so they have teams that are fighting against these games.

They have whole teams internally that are about safety and moderation, that are trying to prevent the cheating of these botnets, but these botnets are still working because it's an arms race. And so these different systems basically create a very complex online social weather system where these different things are happening and you can't basically predict it anymore that these systems kind of create these giant trends that can seem to appear, disappear, but it also can create lasting changes in terms of climate change about what are the fundamental values that we care about.

So what are the human responses to an algostorm? As I mentioned earlier, many will retreat to a smaller, more controlled environment like a home, and engage in genuine human conversations in person, and not bother about the social media weather that is happening. You can imagine a farmer is out in a rural area, who isn't really following all these trends and doesn't know that there's always the outrage about this thing this month, this conflict's happening in that part of the world, and the world's going to be doomed, etc, etc. They're just, "you know what, let's just focus on family, on friends, on what I can control in terms of my career, in terms of my farming, in terms of my small micro environment that actually have agency over."

What I would say is that there is probably the minority and not the majority. Most folks are really unconsciously or semi-consciously following what appears to be the majority opinion. Now, the majority opinion doesn't mean that you're for something. Neither is it because you're against something, but the majority opinion is that you must have an opinion about that thing. That thing could be the US elections, as I mentioned. It could be a political crisis or a military conflict, but everybody has to have an opinion. And, if you think about it, 50 years ago, before the internet appeared, you just had no opinion about it because you're not informed about it. Now, we're not saying that you shouldn't be informed about stuff, but I think, we have to be thoughtful about what are the things that are within our control, and what are the things that we actually have a meaningful say on, and there are many things that we don't have a meaningful say on.

(11:22) Jeremy Au:

The tricky part about following the crowd is that you may think that you're not part of the crowd because you're against it. So for example, it would be like, "Oh, this movie came out, everybody's watching Game of Thrones or this TV series." And so everybody's, "okay, so people have to have an opinion about this." Say, I loved it. I watched it. It was amazing. I can share about the information with it, or I don't want to watch it, but the tricky part is that a crowd is, you have to talk about it. Do you have to talk around it? You can't not have an opinion about watching a show. You're either for it, against it, or you're clueless about it. The existence of that trend plays that game where people are already carved out into different factions.

And some people respond to an algostorm the same way they'll react to a storm, which is that they're going to use that algostorm to push their own agenda, just like how some people who are sailing or surfing will use that weather to push and move faster. They'll align themselves with the trending topics. That is a term that we use in marketing, which is called "trendjacking". It's basically saying, spotting trends before they are new and then following along with it. We all remember that old challenge, the viremia of the ice bucket challenge, which was that, you, dumped, bucket of ice upon yourself and you film yourself doing it and then you would name several other names who should do that challenge and then they themselves would do that challenge as well. That was one of the few episodes that you can imagine where there were different corporations and different people who were trendjacking, they were following the trend, and pushing it to advocate for something and now we see many versions of that kind of piece where corporations and so forth are trying to be on top of that social media conversation trying to be hijacking that trend or in many ways trying to surf that algostorm.

So how will algostorm develop in the future? We already have good examples of that. For example, if you look at the stock markets and something they call flash crashes. So if you look at stock markets, the majority of trading volume is now algorithm-driven. Now, there are obviously many humans who are working as part of the process and they can't be at the desk during an entire time. So what they're doing is that they're setting up algorithms that basically exhibit and execute their training strategy. So they have certain rules, right? If this happens, then do that. If something else happens, then do Z. Algorithms are reflecting human strategies, But there are also high frequency traders who are looking to reduce the arbitrage, right?

So they're saying, if I spot that somebody's selling it for 50 cents here, but on someone else who wants to buy it for 52, I'm going to kind of like, in that microsecond, basically, buy at 50 or pay 51 and then go sell it at 52. But basically, it's like trying to reduce that spread as much as possible so that there's no inefficiency in the market. And so that creates a ton of volume in a market where if you're playing the public stock markets, you're actually, frankly, kind of a minority as a human retail investor. The majority of trading volume right now is being done by bots reflecting humans or bots reflecting certain strategies that are really bot-centric. And so what happens is that now we see the phenomenon of flash crashes, which is what happens when trading happens in a certain way. And sometimes it could be a bug. It could be some sort of political news, but basically what happens is that these rules trigger and they don't really know what to do. And so they often just stop, they pause, they cancel. And so all of the liquidity stops moving because they no longer are trading. And that creates this extreme volatility where, for example, a stock market price can drop to very, very low prices in, say, a matter of seconds to minutes.

And then, the human retail traders are kind of like caught behind that, and then they get called on bings or pings or notifications or messages, and they rush back to their phones to try to figure out what's going on. And then retail investors are caught kind of like lost, kind of confused. And then we often see that there's often maybe a quick recovery, because what happens is that these bots and these humans have to realize Hey, it was an irrational plunge, for example, in a security price. And so now is a great chance to buy back into that security price and to make up for the arbitrage. But that flash crash is also quite interesting because it can really trigger a lot of losses for normal humans who are not using bots. Is a way where the game is now no longer human only, but primarily bot-driven. You can't imagine a bunch of humans rapidly coordinating in a matter of seconds to all agree to sell in conjunction, right? It's interesting where, there are actually players out there that are trying to instigate flash crashes because they know that these bot behaviors do exist. And so there's an interesting arbitrage opportunity, which is, if you can trigger somebody else to, screw up and, cause a crash in the stock market price, then there's an opportunity for you. So it's kind of like game and counter game attacker versus defender.

And so all these things have interacted where, for example, authorities now have circuit breakers. And what that means is that, authorities basically say, if a stock market price drops, below say, 10 or 20% or whatever percentage within a too short a period of time, whether it's 5 minutes, or 10 minutes, or 30 minutes, but, they want to install where the stock times out. So they've removed that stock from trading because they want everybody to come back. They want people to process the news. They want the humans to have enough time to sit down and discuss the news. And then hopefully when they reopen trading on that stock, then the stock will go back to a normal pricing pattern.

(16:17) Jeremy Au:

We also see that stock markets are also trying to regulate and really try to manage whether these high frequency traders are allowed to direct access to the stock market exchange because they don't want to create an unfair advantage that's way too asymmetric, but also because they understand that this is one of the unfortunate side effects of having high frequency trading. That's a useful parallel with the public opinion markets, because now we see that these social media trends are going viral quickly. They have a shelf life of one week or two weeks, and they drive public opinion in waves. And if people feel compelled to react and form opinions on those trending topics. And then within weeks, the digital focus shifts to the next big thing. That viral meme that was really important on Reddit or whatever it was. It's no longer the focus and the previous issue is now forgotten. A lot of this is social or effectively harmless. There was a time when people were debating what a dress looks like, whether a blue color or gold color. There's a big meme, people were all about it. And now I think years onwards, nobody cares about it at all. It's kind of like in a dustbin of internet history. In retrospect, if I think about it, I must have spent a certain number of neurons laughing about it, discussing it with my friends.

And now it's like effectively worthless and has no impact on my life. So this is a pretty harmless viral meme that happened, but other than wasting a lot of people's time. So what we see is that governments are concerned that these things can create a ton of instability, perhaps political or social movements. And so government measures are trying to create an increase of moderation powers and penalties for social media platforms trying to increase control of that. But also, they are willing to go all the way to even shutting down the internet, kind of like an extreme version of a circuit breaker in order to maintain control, whether it's for good or for bad, depending on how you look at it.

So for example, in Singapore, there is the POFMA, the Protection from Online Falsehoods and Manipulation Act, which is, the stated goal is to combat fake news from the perspective of the government. And it very much requires vendors or people who are communicating certain things that they deem to be fake, then they are mandated to, basically, post the government's rebuttal and to articulate that they had been sanctioned to this act.

In Malaysia, you have the Communications and Multimedia Act, CMA, that allows the blocking of harmful online content as defined by the government. In India, in Myanmar and Indonesia, they have enacted internet blackouts. So for example, India has enacted internet blackouts in Jammu and Kashmir. Myanmar has caused internet blackouts to control information. And Indonesia has shut down the internet during political arrests, like the protests in Papua and West Papua in 2019. In 2019, Iran shut down the internet for a whole week. In 2024, Bangladesh also has shut down the internet and then eventually restored mobile internet, but not social media. As you can imagine, there has been criticism about whether internet shutdowns undermine digital rights and the freedom of expression especially when it comes to the internet. So algostorms are not just the algorithms that are designed to manipulate engagement, but also is the fact that humans are interacting in social circles that are much larger than what happened in the past.

(19:09) Jeremy Au:

So what's interesting is that algostorms are not just algorithms that are designed to manipulate your engagement. They are also curators and gatekeepers to the mass of society that are interacting with each other in a way that has never happened before, ever. Humans are biologically designed to talk to themselves, to their friends, to their family, to their tribe, to the village. And only in the past 50 years have people been able to talk to everybody else in the world. When you say something online or post something online, you have a reach of potentially millions or billions of people around the world. And so these algorithms are kind of like massaging and really shaping, but also reflecting a mirror back to individual members of that society, what the rest of society thinks about.

The nuance between an algostorm and a storm of public opinion, of course, is really the degree that it reflects actual human opinion and consensus. If 100% of human society really wanted Hawaiian pizza and algorithms basically helped enhance some of that, but it reflects accurately the opinion of everybody there that 100% wants Hawaiian pizza, then you wouldn't call that an algorithmic storm. It's just a storm of public opinion where everybody suddenly demands Hawaiian pizza. Where the algostorm comes is that, if only 5% actually care about Hawaiian pizza, and everybody else doesn't care, but algostorm makes it such that effectively feels like 95% of society likes Hawaiian pizza from your individual consumption perspective, or that you suddenly have to care about this issue. In other words, when the algorithms are really warping that reflection of human society and public opinion, I think that's really when the algostorm really has its most weirdest, most negative forms because I think when an algorithm tries to push your information that you don't care about and that nobody society cares about is actually very obvious as the algorithm trying to trigger you, I think it's really at its most effective as the algostorm when it somewhat reflects some of the people's opinions of society and makes it the hottest thing of that time.

(21:04) Jeremy Au:

So this algorithmic macro weather that we can't control is that the internet today, if you're looking at the phone, you may feel like the phone is smaller than you, that you are the one who's pushing the buttons on the phone. But actually the phone is actually much larger because it is a giant weather system. And then on the other side of the table, you're not pushing the buttons. It's somebody's algorithm who was designed by an engineer and a business team that's pushing your buttons. So my personal reflection about this is that, whenever I'm reading something on the phone, obviously there's so much power, there's so much information, there's so much learning there, but when I find myself reacting to a trend, feeling, emotion about something, I have to tell myself, does this impact my life? Will I even remember this event or trend in one year? Why am I even reading about this? Do the people I respect care about this set of news? These are critical questions that we have to create that moment of pause before we react to something.

In conclusion, navigating the algostorms of today really requires awareness and critical thinking to take a pause, to really disconnect from time to time, and to reconnect with your individual self. We have to let go of the conventional view that we are the ones pushing the buttons of the phone and getting information from the phone that we get to react to and that we get to process and learn from and take action on. We have to look at the phone as the opposite, that the weather system of a complex society filled with botnets and corporate interests. That weather system is reaching out to the phone and touching our buttons.

We don't want the algorithms to play our emotions like a piano. We have to step back, we have to think about ourselves. We have to seek genuine human interactions. We have to question the digital noise. And we have to remember that these algostorms will only get stronger over time because of better algorithms, more professionalized content, and much more connection of multiple topics with a very deep web of information with a personalized structure where the internet knows you better than you know yourself.

So stay brave, keep calm, and think through whatever that you're feeling whenever you're hearing the internet.

On that note, thank you so much for joining today's episode. Please subscribe to the Brave podcast. Feel free to rate us or leave us a review and follow us at www.bravesea.com for more insights and discussions.