Good Judgement, Bad Judgement — #114

Good morning,

Few of you have probably missed that OpenAI’s CEO and co-founder, Sam Altman, was ousted by the non-profit board on Friday.

The story is still unfolding, and speculations are wild. But it is safe to say that the tech world is in shock, referencing how Steve Jobs was famously ousted from Apple in 1985 after a power struggle with the board of directors.

I’m not the best person to walk you through the details of what has happened and the different reasons that may be behind this decision from the OpenAI board. Instead, I’m linking some resources below.

However … in the last few weeks, building Better Odds has forced me to think about how to act as a start-up in the middle of a technological paradigm shift that is both celebrated and concerning.

Simply put, you want to be an “AI start-up” as often as you don’t like to be an “AI start-up”.

And balancing the two requires a strategy.

We have watched actors like Cambridge Analytica, Theranos, FTX, WeWork and others rise and fall. Charismatic founders have been deceiving investors, board members, and business customers, who no longer take start-ups’ claims at face value. Yes, they all want to benefit from the innovations, too, but not at the risk of losing investments or reputational damage.

You don’t have to believe that AI will either save the world or kill us all. But building a successful company on top of Artificial Intelligence will depend more on judgement than engineering skills.

What values are you feeding into your model? What worldview are you promoting with its computations? How do you balance speed and safety?

One wrong step, and your business is worth nothing.

In one way, this gives me hope.

Better Odds never started because we wanted to use AI, and we don’t consider ourselves an AI start-up. We started with a problem, identified when experiencing again and again how an increasingly crazy world kept throwing companies curveballs, forcing executives to make business-critical decisions without appropriate insights and time to prepare.

Still, AI is helping us efficiently solve this problem. Just like it will help solve almost any technological problem, we have ahead of us. That’s just how a shifting paradigm works. New opportunities should be explored.

But when it comes to technology, my point of view is heavily influenced by Stefan Pettersson, one of Better Odds’ advisors. He keeps telling me … “If you solve a significant problem for your target audience, the technology behind your solution is secondary.”

And that’s how I see AI. It is just one way to solve the problem. We do not depend on it. We apply it where it makes sense, in ways that make sense. We can remove it all, or parts of it, overnight and still have a functioning solution that creates customer value. Yes, running it would be slower, a bit less precise, and more tedious, but it would still solve the core problem.

If anything, the last few days of disorder at Open AI have given me more confidence in this strategy.

Luckily, my judgement has been stress-tested in complex business situations before. And I’ve deliberately chosen advisors who will loudly tell me when I’m an idiot.

So, if my hypothesis is true, this venture should go well.

Time will tell. 🔮

Anna

PS. I’m sending this email from a new newsletter provider today, so if you experience anything weird, I’d love to know.

📍Five Small Things

TALK — Effective Altruism started as a tiny movement among Oxford philosophers over a decade ago. Today, it is repeatedly mentioned in connection to AI development and volatile billionaires. Here’s a TED talk with William MacAskill, one of the individuals behind popularising the movement, explaining what it is all about.

PODCAST — Want to learn more about Sam Bankman-Fried and the fraud at FTX? Listen to this season of Against the Rules with Michael Lewis: The Trial of Sam Bankman-Fried

BOOK — The Coming Wave by Mustafa Suleyman, a co-founder of AI company DeepMind, talks about “the containment problem” — maintaining control over powerful technologies—as the essential challenge of our age.

RESOURCE — Looking for inspiration when making slides? Here’s a curated selection of stunning slide decks.

REPORT — The State of AI Report analyses developments in AI, produced by AI investors Nathan Benaich and Air Street Capital.

🗞️ The News Section

OpenAI CEO Sam Altman Ousted for Lack of Candor in a Surprise Move by its Board

Internet, Artificial Intelligence

OpenAI’s board fired its CEO and co-founder, Sam Altman, on Friday, citing that Altman was not “consistently candid in his communications with the board.” Altman, who co-founded OpenAI in 2015, has led the firm since 2019.

Greg Brockman, OpenAI’s president and another co-founder, was removed from the board but not fired, though he decided to resign due to the news.

OpenAI, which has received $13 billion in investment from Microsoft, is reportedly valued at $86 billion. The company launched ChatGPT a year ago, sparking a generative AI boom.

Altman headed OpenAI’s significant restructuring in 2019, transitioning from a purely non-profit entity to a hybrid model with a for-profit subsidiary, capped at a 100x return for investors. This shift created a complex company structure but enabled a substantial partnership with Microsoft, providing the necessary technology development resources.

In the wake of these departures, Mira Murati, OpenAI’s CTO and a former Tesla employee, steps in as the interim CEO. While OpenAI remains quiet about the details of these changes, Microsoft has expressed continued support for the company under Murati’s leadership.

Here are some sound resources:

• Written: New York Times - OpenAI’s board pushes out Sam Altman, its high-profile CEO

• Podcast: This Week In Startups - BREAKING: Sam Altman fired from OpenAI, chaos ensues! | E1851

• YouTube: Mira Murati, the recently appointed interim CEO and Open AI CTO, was interviewed in an episode of Bloomberg’s The Circuit with Emily Chang this summer. You can also hear early Open AI investor and Microsoft board member Reid Hoffman.

• Podcast: In this episode of Behind the Tech, you can listen to Open AI’s interim CEO and former CTO, Mira Murati, being interviewed by Microsoft’s CTO, Kevin Scott.

• Written: Wired - Who Is Mira Murati, OpenAI’s New Interim CEO?

Osama bin Laden’s ‘Letter to America’ Resurfaces on TikTok, Sparking Concern About Anti-Semitic Content

Internet, Geopolitics

Videos discussing Osama bin Laden’s 2002 manifesto, “Letter to America”, have resurfaced on TikTok recently. The letter was published a year after the September 11, 2001, terrorist attacks planned by al Qaeda.

TikTok creators have now been sharing it in the context of the Israel-Hamas conflict, highlighting bin Laden’s criticism of the United States’ involvement in the Middle East and support for Israel. The origin of the manifesto’s resurgence on TikTok is unclear, and there is no evidence of a coordinated campaign to promote the manifesto on TikTok.

Despite the attention, the videos related to bin Laden’s manifesto did not initially go viral on TikTok, with fewer than 300 videos under the hashtag #lettertoamerica, accumulating around 2 million views. But after a tweet from social media influencer Yashar Ali, the views on the hashtag jumped to 13 million. This has prompted TikTok to intensify its content removal efforts and suppress videos that criticised those endorsing bin Laden’s writings.

The revival of this document on TikTok has drawn attention from lawmakers in Washington and the White House, leading to calls for more action against antisemitic content on the platform. The incident also raised concerns among lawmakers and observers about TikTok potentially radicalising young people and amplifying terrorist writings.

The reappearance of the letter was celebrated on forums used by al-Qaeda supporters, and The Guardian, which published a transcript of the letter in 2002, removed it from their website after it became widely shared on social media.

Top Advertisers Flee X as Elon Musk Faces Backlash Over Antisemitic Post Endorsement

Internet

Major advertisers, including Apple, IBM, and Disney, are withdrawing from X (formerly Twitter) after its owner Elon Musk endorsed an antisemitic conspiracy theory.

On Wednesday, Musk responded to a post on X, which claimed Jewish communities support “dialectical hatred against whites,” by saying, “[y]ou have said the actual truth.” This is part of an antisemitic conspiracy theory suggesting that Jewish people are supporting “hordes of minorities” who are “flooding” into the country to replace white people. This conspiracy theory is the same one that motivated the deadly shooting in 2018 at the Tree of Life synagogue.

Civil rights groups have also reported an increase in antisemitism and hate speech on the platform. For example, Media Matters for America, a left-leaning non-profit, published a report showing how ads from multiple companies appeared next to far-right posts on X. After the report, an X executive declared that the company would no longer monetise the accounts identified by Media Matters and would label specific posts “Sensitive Media.”

X’s CEO Linda Yaccarino said the company has been “extremely clear” about its “efforts to combat antisemitism and discrimination” in a statement on Friday. One of Yaccarino’s primary undertakings has been trying to bring advertisers back to the platform by claiming it’s safe for business.

Musk is instead blaming researchers and political groups. He admitted in September that the company’s US ad business was down 60%, blaming the Anti-Defamation League. Early Saturday, he posted that X Corp “will be filing a thermonuclear lawsuit against Media Matters” and “ALL those who colluded in this fraudulent attack on our company” and argued Media Matters’ report misrepresented the user experience on X.

Israel Demands Meta and TikTok Remove 8,000 Posts for Alleged Promotion of Violence and Terrorism

Internet, Geopolitics

Israel has requested Meta and TikTok to remove over 8,000 posts related to the conflict, citing their promotion of violence, terrorism, and groups like Hamas, according to Forbes. These removal requests increased 10-fold since the war began, with a 94% compliance rate across the major social sites.

Most of these takedown requests targeted Facebook and Instagram, followed by TikTok and other platforms like Twitter and YouTube. Notably, Telegram has not complied with Israel’s demands. In response to the situation, platforms like TikTok and Google’s YouTube actively removed content and channels promoting Hamas, adhering to their policies against material linked to terror groups.

Additionally, TikTok has been facing criticism from Republican politicians in the United States for an apparent increase in pro-Palestine content, which they claimed was an intentional move to sway American youth towards supporting Hamas. TikTok denied these claims in a press release, attributing the trend to the dominant sympathies among its young user base towards Palestine (a claim backed by Gallup polling data). The platform emphasised its algorithm’s neutrality, functioning on a positive feedback loop based on user interactions.

At the same time, individuals with Instagram accounts from or about Palestine report dramatically reduced reach for their content. And journalists reporting from Gaza, without promoting violence or hate, have had their accounts terminated by Instagram. This has made some users suggest that the platforms proactively restrict specific accounts at Israel’s request.

📈 The Insights Section

Paris Agreement at Risk as Governments Fossil Fuel Production Plans Conflicts with Net-Zero Pledges

Climate

Governments globally plan to produce more than double the amount of fossil fuels in 2030 than what would be consistent with limiting warming to 1.5°C.

According to a report supported by the United Nations, plans and projections from governments worldwide indicate an increase in global coal production until 2030 and in global oil and gas production until at least 2050. This creates a global production gap, threatening a well-managed and equitable energy transition.

The planned increase in fossil fuel production conflicts with commitments under the Paris Agreement and clashes with expectations that global demand for coal, oil, and gas will peak this decade. While major producer countries have pledged to achieve net-zero emissions and have launched initiatives to reduce emissions from fossil fuel production, none have committed to reducing coal, oil, and gas production in line with the 1.5°C warming limit.

Instead, countries should aim for a near-total phase-out of coal production and use by 2040 and a combined reduction in oil and gas production and use by three-quarters by 2050 from 2020 levels, at a minimum.

The report also highlights that an equitable transition away from fossil fuel production must consider countries’ differentiated responsibilities and capabilities, with governments having greater transition capacity aiming for more ambitious reductions and helping finance transitions in countries with limited capacities.

TikTok News Consumption Skyrockets Among Adults In the United States, Quadrupling Since 2020

Internet, Journalism

In just three years, the share of adults in the United States who say they regularly get news from TikTok has more than quadrupled, from 3% in 2020 to 14% in 2023. In comparison, news consumption on many other social media sites has declined or remained unchanging in recent years.

According to the Pew Research Center, US adults aged 18 to 29 are the most likely to say they regularly get news on TikTok, with 32% in this age group doing so, which is a higher share than in previous years. In other age groups, the consumption is 15% for those aged 30 to 49, 7% for those aged 50 to 64, and 3% for those aged 65 and older.

Currently, 43% of TikTok users say they regularly get news on the site, up from 33% who said the same in 2022. This means TikTok users are just as likely to get news from TikTok as Facebook users are from Facebook. However, TikTok users are less likely than X (formerly Twitter) users to get news on X.

USDA Unveils Updated 2023 Plant Hardiness Zone Map, Reflecting Warmer Temperatures

Climate

The United States Department of Agriculture (USDA) has released a new Plant Hardiness Zone Map. This map is the national standard for gardeners to determine which plants can survive the coldest winter temperatures in a specific location.

The 2023 map is about 2.5 degrees warmer than the previous map from 2012 across the United States. In the ten years since the map was last updated, half of the country has shifted to a warmer 5-degree half zone, with the central plains and Midwest warming the most and the southwestern United States warming very little.

The plant hardiness map is divided into 13 zones, each representing a 10-degree Fahrenheit range of temperatures, further divided into two half zones of 5-degree ranges. The map has been jointly developed by Oregon State University's PRISM Climate Group and the USDA's Agricultural Research Service. It is frequently used by approximately 80 million American gardeners and growers, and scientists incorporate the plant hardiness zones into research models, including those studying the spread of exotic weeds and insects.

The previous version of the plant hardiness map was released in January 2012, and the 2023 map is more accurate and detailed due to the addition of many new stations and more sophisticated mapping techniques. The new map is based on 30-year averages (1991 to 2020) for the lowest annual winter temperatures within specified locations, while the 2012 edition was based on averages from 1976 to 2005.

The Personality of Founders Key to Startup Success

Business, Psychology

A study published in Nature has revealed that the personality of start-up founders is a critical factor in determining the success of their companies. This research highlights that founders of successful start-ups exhibit distinct personality traits, notably different from the general population. These traits include the 'big five' - openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism.

The study identified six personality types among successful founders: fighters, operators, accomplishers, leaders, engineers, and developers. Using a machine learning algorithm to analyse the Twitter accounts of over 21,000 founder-led companies, the researchers achieved an 82.5% accuracy rate in distinguishing successful entrepreneurs.

A key finding is that a founder's personality is more predictive of start-up success than industry or company age. Start-ups with three or more diverse founders are significantly more likely to succeed. This concept termed the 'Ensemble Theory of Success', emphasises the impact of complementary personalities in a founding team.

The implications of this research are far-reaching, offering insights for entrepreneurs, investors, and policymakers. It also highlights the importance of diverse personality combinations in team dynamics and long-term success across various sectors.

Thank you for reading. I hope you learned something new. ✨

PS. If you got this email from a friend, click here to subscribe.