✋ OpenAI dissolves their AI safety team

ALSO: Stability AI looks to sell

Estimated Read Time: 5 minutes

OpenAI is making more waves… and this time it’s not for good reasons. With Ilya leaving the company and AGI rumors persisting, it seems OpenAI is dead set on releasing AGI before anyone else. No matter the cost.

• 🫢 OpenAI dissolves their AI safety team.

• 👎 The fall of Stability AI.

• 🤔 Apple’s upcoming AI update with OpenAI.

• 🚕 UK creates the first law for driverless-cars.

🫢OpenAI dissolves their AI safety team

What happened: OpenAI has reportedly dissolved their long-term AI safety team only a year after they announced the group.

The details: 

• The AI safety team, also known as the superaligment team, was announced only a year ago.

• The goal of the team was to ensure the safety of future AI systems.

• This happens mere days after Ilya Sutskever and Jan Leike announced their departures from the company.

• Leike believed OpenAI had become obsessed with creating the first AGI (Artificial General Intelligence) at the cost of safety. 

Why it matters: This is concerning. AI is incredibly powerful… and dangerous. And most countries are lagging behind on protective legislation. Companies need internal safety regulations to prevent dangerous AI from being unleashed on the world. It increasingly looks like OpenAI is traveling down a very dark and dangerous road in the name of “progress” and profit.

💁‍♀️ Scarlet Johansson says OpenAI wanted to use her voice. OpenAI is removing one of the voices on ChatGPT because it sounded so similar to her voice. Johansson hired legal councel to find out if it was based on her voice. OpenAI claims it’s another voice actress. Johansson claims OpenAI approached her last year to supply the voice, but she refused and was shocked to hear the voice at GPT-4o’s reveal. Honestly… her voice isn’t super unique. It’s just famous. There’s probably other voice actresses out there with similar voices who could have filled the role. But that’s for the courts to find out.

🇬🇧 The UK opened an office in San Fransisco to analyze the risks of AI. With an AI safety summit kicking off in Seoul soon, the UK (specifically their AI Safety Institute body) is stepping up their efforts by opening an office right at the center of the action. SF is where companies like OpenAI, Google, and Meta build their tech. The UK plans to keep a close eye on these major tech companies. Earlier this month the Institute released Inspect, its first set of tools for testing the safety of foundational AI models. Meanwhile, I’m sure these companies are trying everything they can to NOT get regulated.

😵The fall of Stability AI

What happened: Stability AI has reportedly held discussions with at least one buyer. That’s right… Stability is looking to sell the company. What was once a company evaluated at $1B is now losing tens of millions of dollars.

The details:

• Stability AI is reportedly in talks to sell the company.

• Stability only generated $5M in the first quarter of 2024 and lost over $30M in the first quarter alone.

• They owe around $100M in unpaid bills and debt.

• CEO and founder, Emad Mustaque, recently left the company after multiple failed attempts to raise money.

The history of Stability AI’s Downfall:

• Emad Mustaque founded Stability in 2022 and quickly found success with their open-source AI image generation model, Stable Diffusion.

• In October 2022, Stability was evaluated at $1B.

• The company ran into financial issues in mid-2023 as infrastructure costs continued to climb. Specifically because of the cost of renting the GPUs necessary to process data.

• This is also around the time when Stability started racking up debt with unpaid bills.

• In July 2023 Stability made efforts to raise $500 million in cash and $750 million in computing facilities from investors like Nvidia, Google, and Intel… which fell through.

• In late 2023 investors pressured for Mostaque to step down due to his erratic decision-making and inability to devise a sustainable business plan.

• Around October Stability had less than $5M in the bank.

• Finally in March of 2024, Emad Mustaque stepped down.

Why it matters: This is a sad day for AI. Open source is the solution to big tech’s monopoly on AI. Stability might never recover from this even if they sell to another company and revitalize their structure. This may also mean one less good, updated tool on the marketplace (and Stable Diffusion was very popular).

🤔Apple’s upcoming AI update with OpenAI

What happened: Apple and OpenAI are reportedly planning a major joint announcement on June 10th at WWDC to integrate OpenAI’s model into IOS 18.

The details: 

• Apple is working with OpenAI to bring their AI to IOS 18 and will announce it at WWDC on June 10th.

• Siri will likely be reworked to function closer to OpenAI’s GPT-4o in conversational abilities.

• Some rumored features are summarizing notifications, news articles, and voice memos, as well as AI-based photo editing (more fakery, YAY).

• Apple may possibly revisit building an AI-based search engine.

• Apple’s own AI model is not ready despite hundreds of millions spent on AI R&D.

• Apple has previously attempted to strike a deal with OpenAI, then Google, and OpenAI again. There’s a possibility some Google features may be included anyway.

Why this matters: This could be the biggest shakeup in AI this year. However Apple incorporates AI will tell us the future direction of the tech.

🚕 UK creates first law for driverless-cars

What happened: Self driving cars are officially coming to the UK by 2026 thanks to The Automated Vehicles (AV) Act in the U.K. officially becoming law.

The details:

• The law enables self driving cars to be legal by 2026.

• This is a “groundbreaking” law for self driving cars, being among the first of its kind.

• The law is aimed at increasing safety standards while not taking away a person’s ability to drive for themselves.

• Driverless cars were already legal, though with very harsh restrictions. These laws are meant to compensate for advances in the industry.

• Corporations will now be liable for any mishaps and crashes caused by the self driving cars.

Takeaway: Safety has to be the number one concern here, and it appears to be the case. Liability for corporations also encourages them to take safety seriously. California is also thinking of establishing AV laws, so this could spread to other regions.

🌙 Aliens!

Source: Midjourney

Give us your feedback!

Got feedback? Rate today’s newsletter by clicking below! 

Got more personal feedback? Fill in the box when you rate us and give us your criticism and feedback.

Thank you for reading! 

❤️Share the (ARTIFICIAL) Love!

Got a friend who wants to appear as smart as you? An AI fanatic or someone shaking in their boots of getting replaced? Share us by clicking the link below and you’ll get our love ;)