AI Ethics & Policy Insights

The AI ethics & policy info we should know - simplified. Week of March 4, 2024. Issue 23.

Paige

Mar 08, 2024

ELON MUSK SUES OPEN AI - Elon Musk announced that he is suing OpenAI for deviating from the original mission that was agreed upon when Musk founded the company with Sam Altman and others. Musk claims that the company is no longer in line with the original vision. It should be noted that Musk created a competitor to Open AI. In the course of the lawsuit, he has suggested that Open AI could settle by either merging with Tesla, or changing their name to “ClosedAI”.

MICROSOFT INVESTS IN MISTRAL AI - Microsoft* invested over $16M in French AI startup Mistral AI. Mistral was the talk of the Davos Promenade at the World Economic Forum in January. This is huge for the company, which is less than 1 year old. Mistral is the creator of large language models (LLMs) (making it a competitor of OpenAI’s ChatGPT) and is a huge proponent of open source software. This announcement came just weeks after Google announced a similar partnership with Hugging Face, which is also a French company, and has prioritized open source software and community. It is of note that Google called for the federal government to look into Microsoft’s partnership with Mistral, and claim it is a monopolistic move.

MICROSOFT INVESTS IN FIGURE - It’s a big money week for Microsoft as they continue to spend their dollars investing in the next wave of AI companies. Microsoft invested in the robotics startup Figure, which creates large language models for robotics. Microsoft’s $90M investment puts a significant emphasis on the next wave of AI, which many expect to be humanoid robots.
KLARNA IS GOING BIG ON AI - Swedish fintech company Klarna is going big on AI, and it has some wondering if former employees paid the price. The company, which offers “buy now, pay later” financial services shared that AI is handling 2/3 of customer service chats, which is around 2.3 million conversations. The CEO of the company boasted that AI is doing the work of 700 full time employees. This didn’t land well with the press, as Klarna had laid off that exact number of individuals not long before the CEO began to boast. The CEO claims the matching numbers are just a coincidence.
APPLE RE-FOCUSES ON AI - Apple, which was rumored to be building an electric car, has apparently canceled the plans to re-focus on AI. Bloomberg reported that this is an abrupt change for the company, which has spent decades on the car project. In the AI era, many have wondered when Apple would release new products with AI at the center.
GOOGLE’S GEMINI CHALLENGES - Google can’t catch a break in the world of AI. This was evidenced of late by the major challenges associated with the release of the company’s multi-modal AI, Gemini. When prompted, the model would only put out photos of darker skinned individuals. This highlights bias or “unfairness” in the way the model was either trained, the way the algorithm was designed, or a hyper-active ‘safety’ feature designed to prevent or minimize bias.

*Paige works for GitHub, which is owned by Microsoft.

Deeper Dive

You’ll notice that this week we’re without a deeper dive. That’s because the topic I had planned needed more research and fact checking before it was ready for publication. I write all of my own deep dives (I don’t use AI to do that work) so it occasionally takes a little longer. Instead I’ve linked to older newsletters with deeper dives that you might enjoy catching up on.

A primer on approaches to AI regulation
The dark side of generative AI
Davos (World Economic Forum) AI review
Multi-modal AI is the future
AI replacing humans in the work place
Women in AI - are they hard to find?

AI Policy Watch

Nebraska Lawmakers & Regulating AI Political Ads

Nebraska has entered the AI chat. Lawmakers in Nebraska are debating whether they should require clear disclosure of use of AI in political ads airing across the state. Senator John Cavanaugh drafted LB1203, stating that “Nebraskans deserve transparency.” This is exceptionally fitting as “transparency” is a key principle of responsible AI.

New Jersey Schools Join Others in Creating AI Policy

A school district in New Jersey is joining others in creating AI policy that it says will put the district “ahead of the curve”. With AI’s continual rise and ever-expanding use cases, schools are looking to identify ways the technology can benefit students, and where to put some boundaries. Expect to see more of these policies throughout 2024. (I anticipate a HUGE rise over the summer, in prep for the 2025/2025 school year.)

EU AI Act Vote is March 13th

The EU AI Act has been in conversation, creation and shaping for several years. Next week, the EU Parliament is expected to pass the EU AI Act. While the Act won’t largely go into effect until 2026, some of the provisions will begin in 2024. According to Bloomberg, “The Act would regulate different uses of AI based on risk. It includes prohibitions on certain use cases—such as “emotion recognition” systems in the workplace—which carry the highest fines: 7% of global revenue or 35 million euros (about $38 million), whichever is higher.” The EU AI Act is a first-of-its-kind regulation that is sure to have an impact all over the world.

Video of the Week

My new series Cooking & Baking with AI is meant to bring levity (get it?) to the topic of AI ethics, which can occasionally get a bit heavy. I ask AI to generate a recipe, and then I make it and judge it!

Worth a Read

Let’s Connect.

Subscribe to my Substack.

Connect with me on LinkedIn.

Subscribe on YouTube.

Follow on TikTok, where I talk about developments in responsible AI.

Email me thepaigelord@gmail.com if you want to connect 1:1.