Good morning. Today, we’ve got a bit of a long read in the form of my breakdown of a new study (backed by Sam Altman) about the impact of universal basic income.

In the future, when an edition is a little longer than normal, I’ll give you a ‘longread’ heads-up at the top.

— Ian Krietzberg, Editor-in-Chief, The Deep View

In today’s newsletter:

• 🌲 AI for Good: Reforestation efforts in Kenya

• 🇨🇳 Nvidia is making a new flagship AI chip

• 📈 Later today, Google and Tesla will report Q2 earnings

• 💰 The results of Sam Altman’s multi-million-dollar UBI study are in

AI for Good: Reforestation efforts in Kenya

In 2022, the Kenyan government unveiled a reforestation program with the goal of planting some 15 billion trees across the country by 2032. Part of this campaign is specifically focused on addressing reforestation of water towers, which refer to forested regions that retain the water that sources rivers across Kenya. 

The Kenyan government partnered with IBM to leverage the company’s geospatial foundation model, which IBM fine-tuned for this project. 

The details: The model is being applied to not only track progress on the effort, but to further target and supercharge the campaign. 

• The Kenyan government first established ‘blocks’ in certain regions, each with specific conservation goals (i.e. pure reforestation, soil rehabilitation, enabling controlled grazing, agro-forestry, or energy transition). 

• It then put these blocks up for ‘adoption’ by community groups, NGOs and the private sector. 

The IBM model is being used to allow the government to first determine the extent of deforestation for each block, then track the progress of efforts made to address that deforestation. 

A little more: Since this application is based on one geospatial foundation model, that same model can be fine-tuned and leveraged for any number of countries (and any number of climate-related goals) with minimal additional compute. 

• The Kenyan government is also using this model to access more accurate carbon sequestration estimates, which it achieves by evaluating the extent that trees and other plants hold — or release — carbon dioxide.

"Using IBM's geospatial AI model to help monitor our reforestation program has been a game-changer,” Kenya’s special envoy on technology, Ambassador Philip Thigo, told the Deep View. “It will help transform this into a shared country-wide effort, unlock carbon financing, and save time that we can reinvest back into other important national priorities." 

Nvidia is making a new flagship AI chip for China

Nvidia, according to a Reuters report, is working on a new flagship AI chip designed specifically for the Chinese market. 

The details: The chip, according to several anonymous sources who spoke to Reuters, is intended to be compatible with U.S. export restrictions. 

• Nvidia will work with Inspur — one of the company’s major Chinese distribution partners — on the launch of the new chip. 

• Shipments of the chip — which has tentatively been named the “B20” — are planned to begin in the second quarter of next year. 

Nvidia declined a request for comment. Shares of Nvidia lifted 4.7% Monday. 

For context: This comes just a few days after Bloomberg reported that President Joe Biden’s administration is considering implementing even tougher restrictions on exports of high-tech U.S. chips to China. 

• These export restrictions were first established in 2022 and extended last year. 

China accounted for 14% of Nvidia’s total data center revenue in fiscal 2024; that number was 19% the year before. 

• Clay, a sales and marketing AI startup, raised $46 million in Series B funding.

• Phaidra, a startup focused on providing greater data center efficiency, raised $12 million in funding.

• DHS Has a DDoS Robot to Disable Internet of Things ‘Booby Traps’ Inside Homes (404 Media).

• Delta cancels hundreds more flights in struggle to recover from Microsoft outage (CNBC).

• Exclusive: Illicit chip flows to Russia seen slowing, but China, Hong Kong remain transshipment hubs (Reuters).

• Reddit teams up with the NFL, NBA to boost revenue (The Information).

• Rivian CEO RJ Scaringe says too many carmakers are copying Tesla (The Verge).

Later today, Google and Tesla will report Q2 earnings

It’s that time of year again. Welcome to Big Tech earnings, 2024 (part II). 

Both Google and Tesla will report second-quarter earnings after the bell today. The remainder of the ‘Magnificent 7’ will report over the coming weeks. 

Here’s where they stand: Analysts expect Google parent Alphabet to post a 14% increase in quarterly revenue. For Tesla, analysts are anticipating the company’s lowest automotive gross margin (of 16.27%) since 2019. 

• For both companies, artificial intelligence will be a major investor focus.

For Google, this will involve an examination of the results of its integration of AI into Search, in addition to a keen bit of attention paid to the success of its cloud growth. For Tesla, this focus takes the form of its self-driving investments, AI products and robotaxi promises. 

• Analysts and investors including Cathie Wood, Wedbush’s Dan Ives and Morgan Stanley’s Adam Jonas have been excited about the growth opportunities afforded Tesla by its investments in artificial intelligence. 

• But at the same time, Elon Musk has made many robotaxi promises that haven’t even come close to hitting the mark.

Ives anticipates that “2Q earnings will be a major positive catalyst for the tech sector.” He said recently that he expects Alphabet and Microsoft to be among the standout performers.  

Invest smarter with Public and AI

If you're looking to leverage AI in your investment strategy, you need to check out Public.

The all-in-one investing platform allows you to build a portfolio of stocks, options, bonds, crypto and more, all while incorporating the latest AI technology — for high-powered performance analysis — to help you achieve your investment goals.

Join Public, and build your primary portfolio with AI-powered insights and analysis.

The results of Sam Altman’s expansive UBI study are in

A key facet of the conversation around artificial intelligence — especially artificial general/super intelligence — has to do with Universal Basic Income (UBI), which can be thought of as a minimum income or allowance provided to the people by the government. 

The idea of a UBI has been touted and pushed by several tech businessmen, prominently including Tesla Chief Elon Musk and OpenAI Chief Sam Altman. Both men have claimed that, since AI is apparently destined to replace the bulk of human workers, we need a UBI. 

• Musk said in 2018: "Universal income will be necessary over time if AI takes over most human jobs."

• Several years ago, Altman said that a basic income is an “obvious conclusion.”

Altman has been working on the question of a UBI for years; an OpenAI-backed research lab called OpenResearch launched in 2015 with the initial goal of more expansively studying the effects of a UBI. Altman is one of two members who sit the lab’s board; he personally donated $14 million to the lab while OpenAI’s nonprofit arm has donated $10 million, according to Forbes. Twitter founder Jack Dorsey has donated $15 million to the lab. 

What happened: In 2019, OpenResearch began a three-year study on basic income, which served as the largest U.S.-based study on the topic. This week, the lab began releasing some of its findings. 

•  Over the three-year run of the study, 1,000 (low-income) participants in Texas and Illinois were given $1,000 per month, no strings attached. A control group of 2,000 participants were given $50 per month. 

• Largely, the cash infusion resulted in more individual flexibility and agency, with participants spending more time on leisure and less time working. 

The results: The basic income caused participants’ total income (excluding the monthly transfers) to fall by about $1,500 per year compared to the control group. The researchers noted a 2.0 percentage point decrease in labor market participation, saying that participants worked about 1.3-1.4 hours fewer per week. 

• The researchers in a separate paper noted that the cash infusion ($12,000 per year) resulted in no health impact among participants. 

• Note: this study is far from the first of its kind. It’s just on the larger end of the scale. 

The problem with UBI: For one, it’s really expensive. Daron Acemoglu, an MIT Professor of Applied Economics, wrote in 2019 that “UBI is a flawed idea, not least because it would be prohibitively expensive unless accompanied by deep cuts to the rest of the safety net.”

• In the U.S., which has a population of roughly 336 million, a UBI of $1,000 per month would cost roughly $4 trillion annually. This is just slightly short of the total revenue ($4.5 trillion) brought in by the federal government in 2023. 

• Acemoglu said that taxes would need to increase by significant margins to pay for a universal basic income which would redistribute country-wide wealth among the poor and wealthy alike. 

It’s worth noting that, of that $4.5 trillion revenue number, fully half was generated by individual income taxes and only 9% by corporate income taxes. 

My problem with the idea of UBI is that it presupposes that AGI is both inevitably possible and is coming soon. In a way, Silicon Valley’s push for UBI is yet another example of the rampant hype that has permeated the industry and the wider world; it is predicated on pseudoscience. 

• I’ve said it often and I’ll say it again: there is no scientific evidence that AGI will ever be possible. It might be decades away, and it might be centuries away. It might never be achieved. 

• Whatever AGI might look like, and whatever timeline it is on, it remains a scientific hypothetical. A possibility, but not a guarantee. 

Moving beyond that, let’s say for argument’s sake that OpenAI achieves a (peer-reviewed) AGI tomorrow, sticks it in legitimately functioning robots and subsequently erases the bulk of the human workforce, thus requiring a UBI. Where is the money for that UBI coming from? 

Since it would require a massive increase in taxes, and since people would no longer have income to be taxed, the company(/companies?) selling AGI would need to cough up to fund that redistribution. 

Time for a thought experiment: if that company’s money to fund a UBI is based on selling AGI, and no one has any non-UBI income to make purchases, who is that company selling its AGI to? I don’t know about you, but on $1,000 per month, I won’t be dropping $20 for ChatGPT, or Netflix, for that matter, or the latest iPhone.

• Play out this thought experiment. It doesn’t work. If people can’t work because there are no jobs, and if they are dependent on a fixed, limited income, the customer base for companies that sell goods/services (every company) will plummet. 

• So yeah, they’ll get cheap labor (or maybe expensive labor, because data centers and robotics), but revenue will likely take a hit and so will profit, again, because of those pesky taxes needed to fund a UBI. 

But perhaps the most important question pertaining to this hypothetical scenario is this: how will these few surviving companies be required to redistribute their profits to the masses? Will they be required? 

This whole idea teeters on the edge of full-blown dystopia. It places corporations in, at least, de-facto control of the country and its people. It requires individual dependence on tech corporations that have demonstrated a care, not for constituents, but for corporate profits. 

It relies too heavily on an expectation of moral goodness that the tech sector has shown again and again does not exist. 

• In this scenario, what happens if the small handful of companies powered by said AGI decide to withhold an individual’s UBI? Or choose to stop paying their UBI tax altogether? 

If companies meant business about UBI, I would feel more confident if they had begun this AI frenzy by actually paying artists and writers for the content they’ve scraped off the internet to train their AI models.

I would also feel more confident if these companies redistributed their massive, multi-billion-dollar profits to help fight climate change in real ways; instead, of course, they are moving further from their climate goals due to their investments in AI. 

I see no evidence that the corporate side of the AI industry has much interest in helping people. 

I don’t expect to see a true workforce-erasing AGI anytime soon. But if we got one tomorrow, I do not believe a UBI would act as a simple silver bullet to that problem. 

I see no way it could feasibly work.

A poll before you go

Thanks for reading today’s edition of The Deep View!

We’ll see you in the next one.

Your view on AI in athletics:

30% of you said the use of AI to help athletes train isn’t a bad idea; 22% think it would be a game-changer. 10% think it’s unnecessary and 16% wouldn’t ever use it.

Something else:

• “I can envision uses in reviewing data to help an athlete improve their performance. I think using AI to select athletes sounds awful. All of this still has me worried about scope creep and ever-increasing invasive uses of AI. Let AI free athletes from doing the dishes so they can train more.”

Not a bad idea:

• “Athletes with money can afford professional trainers. Athletes, and non-athletes alike, without money, can certainly benefit from AI assisting with their training, reminders, and motivation. Anything that helps you become more active is good. Move it or lose it.”

^{*Public disclosure: All investing involves the risk of loss, including loss of principal. Brokerage services for US-listed, registered securities, options and bonds in a self-directed account are offered by Public Investing, Inc., member FINRA & SIPC. Cryptocurrency trading services are offered by Bakkt Crypto Solutions, LLC (NMLS ID 1828849), which is licensed to engage in virtual currency business activity by the NYSDFS. Cryptocurrency is highly speculative, involves a high degree of risk, and has the potential for loss of the entire amount of an investment. Cryptocurrency holdings are not protected by the FDIC or SIPC.}

^{Alpha is an experiment brought to you by Public Holdings, Inc. (“Public”). Alpha is an AI research tool powered by GPT-4, a generative large language model. Alpha is experimental technology and may give inaccurate or inappropriate responses. Output from Alpha should not be construed as investment research or recommendations, and should not serve as the basis for any investment decision. All Alpha output is provided “as is.” Public makes no representations or warranties with respect to the accuracy, completeness, quality, timeliness, or any other characteristic of such output. Your use of Alpha output is at your sole risk. Please independently evaluate and verify the accuracy of any such output for your own use case.}