Fake your way to riches
The deepfakes edition
Took a break to chill last week. I sat under a Banyan tree for 8 straight hours meditating about new income redistribution opportunities (scams and frauds). My dedication sometimes scares me. And out of the meditating came this week’s issue.
You’ll enjoy reading it and you’ll also share it. I’m not even gonna ask you to because it is that good. if there were Oscar’s for scamsters, this piece would win the award for the best screenplay.
See, anybody can be an income redistribution expert (scamster). But being a good income redistribution expert, that’s hard. The one trait that sets average income redistribution experts from the good income redistribution experts is the good ones are always on the hunt for new income redistribution opportunities.
The stock market is a goldmine of income redistribution opportunities. In a previous post, I told you how you could easily make money by getting some journos drunk and get them to publish a piece that Google is acquiring Suzlon. I also told you about all the mistakes amateurs make when insider trading. But these things can be boring because they're old school.
If you've been watching Money Heist and feel like doing something fancy, I've got just the opportunity for you. Now, let's say you wanna make some money quickly. Like we've discussed earlier, robbing banks or stealing from your neighbours is tough. Moreover, it's hard. Why do you have to do all those pointless things when you can quite literally make millions without having to lift your precious butt from your chair.
Now, let's say you've picked the stock market to make money. You can either buy value stocks and hold them for a decade, like a moron and underperform an index fund while chanting Warren Buffett quotes. Or you can be like Elon and make the prices move the way you want them to. Because getting drunk journalists to publish stories about Google is acquiring Suzlon is easy but still requires some work. And if you're wondering if there's an easy way to make stock prices move the way you want to, like Elon, bro, come on, of course, there is, don't insult me.
One of the easiest ways to move stock prices your way is by using deepfake media. In simple terms, deepfakes are manipulated images, audio, and video clips. For example, it could be a video clip of Byju Raveendran announcing that he's in talks to acquire the Ministry of Education. Now obviously, he would never say that. But using artificial intelligence and machine learning technologies, you can create a realistic video of him saying it.
For example, check out this Tom Cruise video which went viral recently on Tik Tok. Now a layperson wouldn’t be able to tell if this was fake or real:
Here’s one more deepfake video where Obama calls Donnie Trump a dipshit:
This opens up a world of possibilities for aspiring income redistributors (scamsters/ frauds). Moreover, with apps like Faceswap, Reface and Avatarify, you can create simple cheapfakes (less sophisticated deepfakes). But if you want to create something hyperrealistic, even that’s way easier than, say, 5 years ago. All the tools you need to create deepfake synthetic media are open source.
Now, so far, deepfake technology has mostly been used to create pornographic videos.
Another key trend we identified is the prominence of non-consensual deepfake pornography, which accounted for 96% of the total deepfake videos online. We also found that the top four websites dedicated to deepfake pornography received more than 134 million views on videos targeting hundreds of female celebrities worldwide. This significant viewership demonstrates a market for websites creating and hosting deepfake pornography, a trend that will continue to grow unless decisive action is taken.
But as an aspiring scamster, deepfakes are a goldmine for you because very few people are thinking of using them in the financial markets. Except for the people at Carnegie Endowment, not many people seem to be bothered, which is a good thing for you. You have the first-mover advantage.
Go big
We all like doing daring things. One easy way to do that is to move the stock price of a big stock, say like Amazon. Go big or go home, right? If you do something big, it might as well be something the world will remember you and for a long time. So how do you go about this? It's always better to keep things simple. Now the sentiment around Amazon is closely linked to Jeff Bezos even though he's no longer the CEO. We can use that to our advantage.
So, here's how we move Amazon stock price. We know that Jeff Bezos is divorced, and he must be really sad. And when people are sad, they have a habit of doing stupid and emotional things. Like streaking naked across Wallmarts, going on a bender, breaking into a zoo and trying to kiss a monkey, or getting a tramp stamp.
Let's go with a sad call because it's easy and believable. It's quite easy to create a fake video of a drunk and sad Jeff Bezos calling his brother. Here's how the conversation could go:
Jeff Bezos: Hello chotu
Mark Bezos: Bhaiyya kaise ho
Yes, Jeff Bezos speaks Hindi
Jeff: Divorce hogaya, mera life khatam hogaya
Mark: Bhaiyya, ye sab hota he. Ek peg Old Monk pee lo, aur Tinder app download karlo
Jeff: Bahut sadness hu me
Mark: Apke ke paas $190 billion he aur Rocket he, mere pass kuch nahi he. Bahar jaake thoda have fun karo
Jeff: Chotu, tu gareeb he?
Mark: Mujhe, I don’t want your paisa
Jeff: Ok, paise nahi dunga. Par, tomorrow Amazon call options aur Vodafone call options buy karlo. Me ek rumour start karta hu ki, Amazon is buying Vodafone. 10-minute ke baad bech dalo, aur mere jaise rich ban dalo
Mark: Bhaiyya, I love you
Look, people do stupid shit after they are divorced, so this is believable. What's more, Jeff is still hurting from the whole leaked messages scandal. So, people will believe this video. So, now that you have the video ready, sell your kidneys and buy Amazon and Vodafone calls. DM this video to some drunk journalist or create a dummy Twitter account and tweet, and your job is done. The stocks will react, you will be rich.
You might think that people don’t fall for such obvious things. You’d be wrong because people are dumb!
This is a deepfake video by a Belgian political party of Donnie asking the Belgians to exit the Paris Climate Accord. They created the video to raise awareness about climate change, but it backfired.
Quite a few people actually believed it and were ranting on social media.
Go smart
Now, the first idea was too big and could easily fail. But going small has advantages because a lot less people track smaller stocks, and you can easily get away by playing with them. Pick a small stock that is already in the news. This requires lesser efforts. Now, every company has auditors to certify the books of accounts and earnings. These people are super important. Most often than not, whenever there are auditor changes, the odds are there are red flags. Stocks like Vakrangee, Manpasand etc., which fell by 80-90% of the peaks, had auditor resignations. We can use this to our advantage.
And the other advantage is these auditors don’t interact or speak publicly, which means nobody knows how they look or sound. You could potentially create a fake audio clip where an auditor says a negative thing about a company, and you can leak it. You could buy the company’s stock in advance, leak the audio and become Jeff Bezos rich.
This will work because nobody knows how the auditor would sound. By the time the news is verified, you’d have become rich.
Go long
Or, you could play the long game. With video & audio leaks, they're a one-time thing. You have to create fake things continuously, and you can't do it frequently because nobody will believe it. But there's another way. Thanks to advancements in deep learning and natural language processing technologies, today, you have tools like GPT-3 and Tensorflow which are capable of automatically generating human-like text.
GPT-3’s samples are not just close to human level: they are creative, witty, deep, meta, and often beautiful. They demonstrate an ability to handle abstractions, like style parodies, I have not seen in GPT-2 at all. Chatting with GPT-3 feels uncannily like chatting with a human. I was impressed by the results reported in the GPT-3 paper, and after spending a week trying it out, I remain impressed.
So, it's quite simple. Set up a blog and a Twitter account with a fake identity like Pankaj Stonkwala. Now, you don't have to waste your time tweeting anything. You can use automated text generators to write and tweet stuff. Here's an example of how I generated a realistic market action article with just a simple phrase on DeepAi.
Here’s one more such tool called Talk to Transformer by Adam King powered by OpenAi.
Now, if you need this to be even more believable, you can train your own model using all the publicly available tweets and finance-related articles. Create a bot and run an automated Twitter account. You can even train your model to reply to tweets. Here's an example from Lex Friedman:
It's been done before. Spy agencies routinely create authentic-looking profiles on Twitter and LinkedIn. But you have to be careful your bot doesn't turn into a racist asshole like the Tay Bot by Microsoft. Now, once your bot is ready and you're active on Twitter, you'll inevitably build a following. Once you have enough followers, you can start tweeting about stocks over a period of time. If you have a big enough following, anytime you mention a stock, odds are people are going to react. You can use this to move prices and make money continuously.
It also need not be a finance specific account. You can even create a popular character like this Twitter account, for example. What you’re seeing is a young Japanese kid who’s into bikes and has 33,000 followers. But it’s a fake person. A 50-year-old dude runs this account.
Now imagine you build up such a popular account, and over a period of time, you piggyback on its popularity and start tweeting about stocks to move them. This is the easiest thing in the world. You don't even have to work hard either. For example, this account has personas generated by deepfake tech, and these people don't exist. You can use these to train your models and to set up your Twitter account.
Here’s a scarily believable deepfake of Zuckerberg:
Customer satisfaction is important, even for illegal things
Customer satisfaction is important. No other brand obsesses over delighting customers as much as Amazon does. Here's an excerpt from Amazon's 2016 shareholder letter:
There are many ways to center a business. You can be competitor focused, you can be product focused, you can be technology focused, you can be business model focused, and there are more. But in my view, obsessive customer focus is by far the most protective of Day 1 vitality.
Why? There are many advantages to a customer-centric approach, but here’s the big one: customers are always beautifully, wonderfully dissatisfied, even when they report being happy and business is great. Even when they don’t yet know it, customers want something better, and your desire to delight customers will drive you to invent on their behalf. No customer ever asked Amazon to create the Prime membership program, but it sure turns out they wanted it, and I could give you many such examples.
Staying in Day 1 requires you to experiment patiently, accept failures, plant seeds, protect saplings, and double down when you see customer delight. A customer-obsessed culture best creates the conditions where all of that can happen.
This is important not just for legal businesses but also for illegal businesses, as this example shows. Conti is a notorious ransomware gang that operates a ransomware-as-a-service (RaaS). The way they operate is they take care of the maintenance of the software while their affiliates do the hacking and ransoming. In return, Conti takes 20-30% of the earnings, and the rest goes to the affiliates.
In such a model, it’s important that the affiliates are happy. Otherwise, nobody makes any money. It looks like the Conti gang was greedy in paying the affiliates. One of the affiliates got pissed off and leaked the gang’s training material and other important technical details,
The affiliate said they posted the material as he was only paid $1,500 as part of an attack, while the rest of the team are making millions and promising big payouts after a victim pays a ransom. “I merge you their ip-address of cobalt servers and type of training materials. 1500 $ yes, of course, they recruit suckers and divide the money among themselves, and the boys are fed with what they will let them know when the victim pays,” the affiliate posted to a popular Russian-speaking hacking forum. Attached to the above post are images of Cobalt Strike beacon configurations that contain the IP addresses for command and control servers used by the ransomware gang.
Never take your customers for granted, even if you're scamming.
Adventures in cryptoland
Permanent disablement by a thousand cuts
The biggest risk in crypto, apart from losing all your money in a shitcoin rug pull, of course, is regulation. Globally every country has struggled with this, including India. The regulatory attempts have been a bit like standing on the ledge of a 50-floor building while being drunk and pissing in the wind and trying not to fall.
The US is no different, and they’ve been making a lot of noise about it. The US govt figured that while they figure out how to regulate Ponzi schemes, I mean crypto, they might as well make some money off it. Last week, the US senators introduced a $1 trillion infrastructure bill. And in the bill, they included a provision for the reporting of crypto transfers to tax them. The goal was to collect $28 billion from taxing these transactions.
But the provision had a liberal definition of “broker” in the current version of the bill, which includes intermediaries like miners and developers apart from actual brokers and exchanges like Coinbase. This predictably caused uproar in the crypto circles:
In its present form, the provision would broaden the definition of a “broker” to any entity within the cryptocurrency industry that facilitates the transfer of digital currencies for another person. This could include miners, hardware and software developers, and other parties that help facilitate a transaction, but don’t participate in the transaction, the Blockchain Association, Coin Center, Coinbase, Ribbit Capital and Square said in a joint statement earlier this week.
And given the dysfunctional US legislative system, there was a flurry of proposed amendments to fix this amidst some strong lobbying from the crypto industry. So much for crypto being the solution to the centralised financial system and the big bad Fed, money printing, and suits. There’s some delicious irony in here somewhere, but I can’t quite put my finger on it. Maybe because it’s decentralized? Wait, are decentralized things irony proof too?
Regardless of whether the initial proposal was due to crypto illiteracy or a carefully thought out ploy to reign in the industry, the path for crypto will get a lot thornier. Nobody wants to kill the industry, but they don't want it to grow big either. So the regulators will keep hitting the industry in the knee with a hockey bat with just enough force to hurt it but not permanently disable it.
Adding to this, Gary Gensler, the new SEC chairman, is calling for regulations too:
Right now, we just don’t have enough investor protection in crypto. Frankly, at this time, it’s more like the Wild West. This asset class is rife with fraud, scams, and abuse in certain applications.
There’s a great deal of hype and spin about how crypto assets work. In many cases, investors aren’t able to get rigorous, balanced, and complete information.
If we don’t address these issues, I worry a lot of people will be hurt.In my view, the legislative priority should center on crypto trading, lending, and DeFi platforms. Regulators would benefit from additional plenary authority to write rules for and attach guardrails to crypto trading and lending.
To be or not Binance
Things just keep getting worse and worse for Binance, the largest crypto exchange in the world. On the one hand, it’s facing a mega regulatory crackdown across the world. Just In the last few months:
Several countries like the UK, Italy, Japan and Canada have either banned Binance or issued advisories.
Major banks stopped working with Binance
Changpeng Zhao even offered to resign and abide by rules in the aftermath.
Authorities in India, Malaysia and Thailand, announced probes over money laundering and illegal operations
It stopped offering derivatives in Italy, Germany, and Netherlands.
On Aug 6th, it stopped opening new accounts in Hong Kong and
On 7th, Brain Brooks, the 4-month-old CEO of Binance US, quit.
On the other hand, there is speculation on social media over its registration status, whether the exchange is viable and also whether Binance deliberately crashed the platform not to pay claims on May 19th:
Crypto reads for the week
I had written about how you could be Warren Buffett Rich. All you had to do was to create a new crypto token called “APUNBUFFETT” coin, pump it in the media and on Telegram groups and dump it. The New York Times had a similar piece on just how easy it is to launch a shit token, scam people and make money. I’m wondering if this NYT piece marks the top for crypto
Good reads
Caught in a value-buying trap: why retail investors often get it wrong
Are ‘Best Ideas’ Portfolios Really the Antidote for Active Managers?
It’s not just Robinhood. Money-hungry stock exchanges are under scrutiny too.
How gambling swallowed sports media
Very Insightful!