Fascinating analysis of AI’s effects on freelancer hiring demand: everyone wants a chatbot

Bloomberry Jobs Report

tldr: writing, translation, & customer service jobs declined; everyone wants a chatbot, apparently

A few excerpts:

To my pleasant surprise, most of the job categories actually had an increase in the number of jobs since ChatGPT was released, with the exception of 3 categories that had large declines in jobs.

The above is important to keep in mind for AI luddite doomsayers; as the history of technology has repeatedly shown, though there may be short term disruption and harm to labor, the long term trend is that technology = more (newer and better) jobs

I was surprised to see graphics design, video editing/production, and even software development jobs go up, given all the anecdotal stories we’ve been hearing about people using ChatGPT to generate code, illustrations and even full featured videos.

And surprising on first blush, although perhaps logical if you assume good general purpose AI has reduced the need for specialized ML:

While there was a initial increase in data annotation jobs, the # of data annotation jobs have been fairly flat in the past 10 months. And the # of machine learning jobs has actually decreased a bit since ChatGPT was released.

Finally, I guess everyone loves to chat:

The most popular use case, by far for AI right now is in developing chatbots

Full article: https://bloomberry.com/i-analyzed-5m-freelancing-jobs-to-see-what-jobs-are-being-replaced-by-ai/

Random facts — things I learned (Jan 12 2024) — Data about the past is truth, Data about the future is religion (Auren Hoffman)

Prior editions

The most important takehome is that tokens are not equity, but are more similar to paid API keys. Nevertheless, they may represent a >1000X improvement in the time-to-liquidity and a >100X improvement in the size of the buyer base relative to traditional means for US technology financing — like a Kickstarter on steroids

But entertainment is one market, the market for attention, and one platform is ahead of everyone else in harvesting the commodity: TikTok. The Chinese (and it is Chinese) company had the most ascendant platform in history until OpenAI, and it remains the frame through which Western youth perceive the world

The lesson to be learned from this is that it is often undesirable to go for the right thing first. It is better to get half of the right thing available so that it spreads like a virus. Once people are hooked on it, take the time to improve it to 90% of the right thing.

Like many people, I adhere to a religion that gives me moral guidance. The practice of it is just good sense—it keeps you out of prison and safely hidden in the crowd.

By approaching life with full consciousness, with vitality and intensity, by becoming the masters of our absurd fate — this is how we answer the question of suicide, how we defy futility and establish what it means to live.

The fun lies in tweaking things until they fit just right. Nothing can beat the feeling that happens after that.
And sometimes all it really takes is tweaking just one single thing to go from same same but different, to same same but right

The greater the doubt, the greater the dopamine

Our current “multiplayer” experiences draw too much attention to the multiplayer-ness. The other people around you demand attention. They move. They flash. They point to exactly what they’re focused on, drawing you away from your own focal point. We are missing out on a fuzzier, softer sense of the shared web.

“Memes are fun and memes are also something to come together around. Speculating on the popularity of memes and their staying power is no different than any other form of speculation… I’ve decided that I am going to stop ignoring and dismissing meme investing and start trying to understand it better. I think it is not something that is going away anytime soon and may turn into something even more interesting.” — Fred Wilson, AVC

Crypto is the modern version of the long emerging markets trade. As an industry, it will see the most relative capital inflows, budding innovation and has an innately global footprint. It provides a fiat currency debasement hedge (Bitcoin), new application networks akin to the internet (smart contract blockchains like Ethereum and Solana) and the fastest growing population. You can compare it to any individual sovereign, country or financial market and nothing beats it.

My whole North Star is how do I continue being able to learn, synthesize and teach forever. – Ali Abdaal

You can’t tell kids to follow their dreams when their dreams suck. – Palmer Luckey

Kardashev Type I: capable of controlling the entire energy from its planet
Kardashev Type II: capable of controlling the entire energy of its host star and travels through the solar system
Kardashev Type III: capable of controlling the energy at the scale of its entire galaxy

The barbaric life that I live, that you have to live, the almost obsession that you must have to be great, you can’t put that sh*t in a f*cking book, bro.” – David Goggins

Auren Hoffman:
Data about past = truth
Data about future = religion

Think ism is a disease. We can’t solve problems by thinking about them. You have to do and try things – Kevin Kelly

Price is the maximally compressed signal of economically relevant information.

New study found that tea consumption is associated with attenuation of biological aging.
People who consistently drank >4 cups a day aged 40% slower than those who drank none.

I’ve heard old-timers say that the fifth set has nothing to do with tennis. It’s true. The fifth set is about emotion and conditioning. Slowly I leave my body. Nice knowing you, body. I’ve had several out-of-body experiences over my career, but this one is healthy. I trust my skill, and I step out of its way.

The group is the basic user class for the tools we need today as a society, yet few pieces of software allow the squad as a whole to produce cooperatively and generate wealth together. To fully realize SQUAD CULTURE this must change

Startup valuation is literally a made up number

Having an internet audience is still wildly underpriced

Build products that no one asks for but everyone wants

Staying in conversation mode helps OpenAI thwart your goals. Remember that you’re not having a conversation with a sentient bot, you’re editing a shared document with a robotic LLM

Being entitled is the ugliest thing a person can be – Nolan Bushnell

Theodore Roosevelt: “Far and away the best prize that life has to offer is the chance to work hard at work worth doing.”

“Generalized trust” or “meta-trust” is “trust that whatever issues might arise between us, we can talk about things in a way that is workable for both of us and leads to issues getting resolved to our mutual satisfaction in good time.”

Money is the stored time and energy of other people

Offline, in the real world, OpenAI have designed a fictional character named ChatGPT who is supposed to have certain attributes and personality traits: it’s “helpful”, “honest”, and “truthful”, it is knowledgeable, it admits when it’s made a mistake, it tends towards relatively brief answers, and it adheres to some set of content policies

Since at least the time of peripatetic Greek philosophers, many other writers have discovered a deep, intuitive connection between walking, thinking, and writing. “How vain it is to sit down to write when you have not stood up to live!” Henry David Thoreau penned in his journal. “Methinks that the moment my legs begin to move, my thoughts begin to flow.” Thomas DeQuincey has calculated that William Wordsworth—whose poetry is filled with tramps up mountains, through forests, and along public roads—walked as many as a hundred and eighty thousand miles in his lifetime, which comes to an average of six and a half miles a day starting from age five

When Claude Shannon worked out the math, he found something very surprising: The formula for noise in an information system was identical to the formula for entropy in thermodynamics.

Our thesis at Variant is that the next generation of internet networks will turn users into owners—specifically asset owners. The internet enabled everyone to become a publisher, and similarly, crypto enables everyone to become an asset owner, and therefore, an investor. You don’t need capital to invest, you can invest your time or work by producing art, running machines, or doing physical work.

Resistance will unfailingly point to true North — meaning that calling or action it most wants to stop us from doing. We can use this. We can use it as a compass. We can navigate by Resistance, letting it guide us to that calling or action that we must follow before all others. Rule of thumb: The more important a call or action is to our soul’s evolution, the more Resistance we will feel toward pursuing it. – Steven Pressfield

Given that Japan is the largest holder of US Treasury bonds and the largest international creditor as viewed by its net international investment position, actions of the Japanese private sector could place enormous upward pressure on 10-year plus maturity US Treasury bond yields.
This is data from the IMF which estimates Japan’s net international investment position is a positive $3.3 trillion

The US is now spending materially more money on the national interest payment compared to our defense budget.
That is quite a feat considering we are handing out weapons and equipment around the world like it is candy, so our allies can fight our proxy wars.

Thich Nhat Hanh
-If you see a person and don’t also see his society, education, ancestors, culture, and environment, you have not really seen that person. Instead, you have been taken in by the sign of that person, the outward appearance of a separate self.
-We already are what we want to become. We don’t have to become someone else. All we have to do is be ourselves, fully and authentically. We don’t have to run after anything. We already contain the whole cosmos.

In the traditional art world, each niche subgenre of art (Andy Warhol, post WWII African art, etc.) ends up being cornered by a handful of collectors. That is also happening to the digital art market right now. In 2023 there were around six institutional buyers who spent at least seven figures buying all the grail digital art pieces.

“When you throw the impossible at them, that’s when they get excited” – Hans Zimmer

“Always want more but never be greedy” – Giannis

I want to live in a world where building a humanoid robot at home is like building a PC today

stop thinking of time as an abstraction: in reality, beginning the minute you are born, time is all you have. It is your only true commodity…To waste your time in battles not of your choosing is stupidity of the highest order.

TIL the black hole, TON 618, is so big (66 billion solar masses), a new term was invented to describe it: ultramassive black hole

Never discriminate as to whom you study and whom you trust. Never trust anyone completely and study everyone, including friends and loved ones.

Crypto is the machine’s body.
And AI is the brain that enters into it.

Having averted disaster, I’m suddenly loose, happy. It’s so typical in sports. You hang by a thread above a bottomless pit. You stare death in the face. Then your opponent, or life, spares you, and you feel so blessed that you play with abandon.

Could you make a “Large Solar Flares and Sunspots Model” (LSFASM) and learn to talk to the Sun and ask it where it might flare up next? How about a Large Oceanic Model that allows ships to talk to ocean currents? Or a Large History Model that works as a Prime Radiant for Asimovian psychohistory? Maybe a Large Climate Model constructed out of weather data can talk to us and supply strategies for climate change?

As the bicycle rolled across the world, critics followed, calling it dangerous because it startled horses and unbecoming for women, who had to wear traditionally male attire to ride (‘bloomers.’)
While some physicians argued cycling was healthy others linked it to insanity, deformities of the spine, face and even a cause for appendicitis. One insurance company even refused to insure avid bicycle riders, while one army recruitment office rejected applicants who were avid cyclists because it was assumed they had a weakened ‘bicycle heart.’

If an idiot were to tell you the same story every day for a year, you would end by believing it.
— Horace Mann

The Master said, “The firm, the enduring, the simple, and the modest are near to virtue.”

Nobody has the right to not be offended. That right doesn’t exist in any declaration I have ever read. ~ Salman Rushdie

Once a government is committed to the principle of silencing the voice of opposition, it has only one way to go, and that is down the path of increasingly repressive measures, until it becomes a source of terror to all its citizens and creates a country where everyone lives in fear.
~ Harry S. Truman


I stopped drinking caffeine this year.
It was incredibly difficult to do.
I used to drink a large Starbucks iced coffee each morning & a medium sized coffee mid-day.
That is ~ 300mg of caffeine.
How did I stop? — I went cold turkey one day. It really sucked. I had headaches, I was irritable, and generally felt horrible for about 10 days.
But the benefits made it all worth it. Here is what I’ve seen so far:
– More energy during the day
– Better sleep at night
– More hydrated skin
– Elevated mood
If you’re looking for something that could improve your life in 2024, highly recommend trying to cut out the caffeine drug.

Start or end every day with writing about your life. There’s always something buried underneath the to do list in your head, something you didn’t realise you felt, that when written down, will make everything clearer.

When thinking about new cities in the United States, Las Vegas comes to mind as the most populous city founded in the 1900’s. Since its founding, it has benefited from having a different set of rules than other places. With its legalized gambling, tolerated prostitution and mob activity historically funneling money into and around Vegas, it benefited massively from having different rulesets. But it is not exactly the type of example that inspires policymakers in other jurisdictions

After studying AI and attempting to model various types of financial decision making algorithmically, I realized that humans make decisions very much the same way that modern search engines do. We have vast stores of data — the experiences we’ve encountered in our lives — and we use very simple algorithms to make predictions and decide on actions. I recollect what happened in my past circumstances, and based upon that history of evidence, I’m going to extrapolate the likely outcome of the current situation and choose the best course of action.

The English language will increase its dominance in an AI world

Language is itself a technology, and like many technologies, it exhibits a classic network effect: each additional speaker of a language increases that language’s utility for all other speakers. The more “users” who speak and write English, the more valuable it is to know and use English in just about all affairs.

One obvious example is in software programming. Though there is a lot of symbolic and mathematical notation in programming languages, most would agree that English is head-and-shoulders more valuable to know (relative to the 2nd or 3rd most popular language) if you want to be a good programmer. It’s better for troubleshooting, for reading documentation, for scouring StackOverflow for copy-paste code, and now for getting ChatGPT or CoPilot to write code for you.

My belief is that as AI proliferates, English will only increase its lead. English is already in the lead with 1.4B speakers (though this number varies significantly depending on how you measure fluency), and Mandarin Chinese is second at 1.1B.

Why?

AI models need data. English comprises a majority of the available online training data. It helps that the largest economy in the world (the US) and the most populous country in the world (India, which depending on your reference, surpassed China’s population this year), are both English markets.

The largest content generating internet platforms — from Google to Facebook to Twitter to Wikipedia and on and on — are dominated by English speakers. An AI model’s output quality is directly correlated with the quantity of its training data, and there is simply more English data available than any other language, including Mandarin Chinese. Thus GPT4 and LLAMA and so forth are “smartest” in English.

There are multiple reasons why Mandarin Chinese lags behind, beyond just the fact that the breakthrough innovations in AI research and productization happened first in the US and UK. Among these reasons are the Great Firewall, the highly regulated and controlled nature of Chinese data, and China’s pervasive digital censorship (For example, there are more than 500 words alone that can’t be used on many Chinese UGC websites because they are perceived as unfavorable nicknames for President Xi Jinping)

Thus Chinese online training data lags English in both quantity and likely quality. There are also some reasons related to the languages themselves, where English is a more explicit language and Chinese more contextual.

English’s initial data lead is a self-reinforcing feedback loop — the more that people use English to interact with services like CharacterAI and ChatGPT, the more data the LLMs have to refine and improve (in English). Leaving other languages in the dust, especially long tail ones like Icelandic or Khmer.

As AI agents increasingly interact with each other, I’m guessing they will develop their own unique protocols for AI-to-AI communication. Not dissimilar to how computers communicate via highly structured network requests, only more complex and perhaps unique. AI will eventually create its own AI lingua franca. However, it’s also necessary that some human-readable component be built into this AI-ese (because at a minimum, developers will want to know where to debug and fix errors). English will likely be chosen for that AI-to-AI interface.

Of course, AI is an amazing and broad innovation that will benefit speakers of all languages. It will help to preserve and distribute rarer languages, and enable faster and better language-to-language education and translation. Whether you speak Vietnamese or Icelandic, there will be an AI model for you. I’m simply arguing that these secondary languages won’t be anywhere NEAR as good as the leading English models, and I would venture that even if English isn’t your first or even second language, you will probably still get better results using broken English to interact with ChatGPT than, say, French.

I could be very wrong here. As with any emerging technology, second and third order effects are by their nature unpredictable and chaotic. And the technology still has a long way to evolve and mature. Let’s see how it all plays out. I’m especially curious about what kinds of AI-to-AI communications will emerge, whether exposed through a human-readable interface or otherwise.

Ok that’s it, over and out good sers and madams! OpenAI wow!

“AI has a far greater likelihood of being dangerous in a government’s control”

Complex adaptive systems have like 20 different inputs (that you know of) and 4 different Chesterton’s Fences in them that will produce about 14 different 2nd-order effects, 21 different 3rd-order effects, and 11 different 4th-order effects you didn’t even know could happen.

Models for CAS that I disrespect: the economy, the climate, and any ecosystem with a massive amount of independent and dependent inputs and outputs. Before 2020 I wouldn’t have bucketed epidemiology in here, but it’s a worthy new addition to the “you really don’t know what you’re talking about, do you?” team.

And:

AI has a far greater likelihood of being dangerous in a government’s control than it does in the hands of genuinely brilliant engineers trying to create a transcendent technology for humanity. e/acc.

Thought provoking.

Who’s in the loop? How humans create AI that then creates itself

If you think about the approximate lifecycle of AI that’s being built today, it goes something like this:

1. Write algorithms (eg, neural nets)
2. Scrape data (eg, text and images)
3. Train (1) algorithms on (2) scraped data to create models (eg, GPT-4, Stable Diffusion)
4. Use human feedback (eg, RLHF) to fine tune (3) models – including addition of explicit rules / handicaps to prevent abuse
5. Build products using those (4) fine tuned models – both end-user products (like MidJourney) and API endpoints (like OpenAI’s API)
6. Let users do things with the (5) products (eg, write essays, suggest code, translate languages). Inputs > Outputs
7. Users and AI owners then evaluate the (6) results against objectives like profitability, usefulness, controllability, etc. Based on these evaluations, steps (1) through (6) are further refined and rebuilt and improved

Each of those steps initially involved humans. Many humans doing many things. Humans wrote the math and code that went into the machine learning algorithms. Humans wrote the scripts and tools that scraped the data. Etc.

And very steadily, very incrementally, very interestingly, humans are automating and removing themselves from each of those steps.

AI agents are one example of this. Self-directed AI agents can take roughly defined goals and execute multi-step action plans, removing humans from steps (6) and (7).

Data scraping is mostly automated (2). And I think AI and automation can already do much of the cleaning and labeling (eg, ETL), in ways that are better cheaper faster than humans.

AI is being taught how to write and train its own algorithms (steps 1 and 3).

I’m not sure about state of AI art for steps (4) and (5). Step 4 (human feedback) seems hardest because, well, ipso facto. But there are early signs “human feedback” is not all that unique, whether using AI to generate synthetic data, or to perform tasks by “acting like humans” (eg, acting like a therapist), or labeling images, etc.

Step (5) is definitely within reach, given all the viral Twitter threads we’ve seen where AI can build websites and apps and debug code.

So eventually we’ll have AI that can do most if not all of steps 1-7. AI that can write itself, train itself, go out and do stuff in the world, evaluate how well it’s done, and then improve on all of the above. All at digital speed, scale, and with incremental costs falling to zero.

Truly something to behold. And in that world, where will humans be most useful, if anywhere?

Just a fascinating thought experiment, is all. 🧐🧐🧐

These times are only gettin’ weirder.