🚶Amazon walks back on cashier-less stores

ALSO: Stability AI has serious financial problems

Estimated Read Time: 5 minutes

Amazon continues to struggle in the retail market as their cashier-less AI tech struggles to remain viable. And Stability’s dirty laundry comes out as reports reveal how broke they are.

• Amazon removes cashier-less feature in stores.

• Stability’s CEO abandoned ship as money ran out.

• OpenAI updates their custom model.

• Meta doesn’t like interracial couples.

🚶Amazon walks back on cashier-less stores

What happened: Well that didn’t last long. Amazon’s walking back their “cashier-less” Just Walk Out experience in larger US stores. This will remain for smaller corner stores though. And you wanna know a secret? Amazon’s tech wasn’t even AI powered…

The details: 

• Amazon is removing Just Walk Out checkout experience from larger US stores. 

• Amazon has faced severe challenges in the retail space and closed several retail stores recently.

• Dash Carts will replace the Just Walk Out experience, allowing users to scan items as they shop.

• UK stores will remain as they are.

• Amazon's evolving from camera-based to RFID and palm-scanning tech for purchases.

But that’s not all:

Amazon’s Just Walk Out feature was touted as this huge tech revolution that would change the way we shop. Instead… it’s powered by 1,000 employees in India who watch and review customer purchases. Yes the AI system did a lot of the lifting, but human manpower was still needed. 🤦 What a show.

Why it matters: This was still a leap in tech, even if Amazon overplayed their hand. Gotta wonder if the desire to close these stores is due to theft… or manpower costs. But the way we checkout is changing, from credit cards to phone scans, and now to biometrics and camera recognition. What do you think the next step will be?

As of market close April 07.

Users are reporting Claude 3 has become more censored since becoming popular. While Claude was lauded for it’s intelligence and willingness to cooperate, it seems popularity has forced Anthropic to censor Claude. Things like political topics and giving “definitive” answers are becoming harder to accomplish. It’s not at ChatGPT’s level yet though (you literally have to brainwash ChatGPT sometimes). This may be a problem with all AI tech when it becomes more popular.

Saudi Arabia is planning a $40B push into AI. This follows substantial investments to diversify their economy past oil alone. Saudi's Public Investment Fund previously invested in SoftBank's Vision Fund, which supports AI and tech companies. Yes another outrageous sum of money thrown at AI.

Youtube CEO, Neal Mohan, stated that if OpenAI trains Sora (their video generator) on Youtube, it’ll be against Youtube’s terms of service. But when asked how this would violate their terms (and what the consequences would be) the CEO dodged the question. Well done, Mohan. I have no doubt Sora is training off of Youtube. But I’m sure Google would rather make money by licensing the Youtube platform officially for AI training (or maybe it wouldn’t be worth it?) just like Reddit and many popular newspapers have recently done.

🪙Stability ran out of cash before the CEO left

What happened: Stability just lost their CEO, Emad Mostaque, and it turns out the company was in serious trouble. Unpaid bills, expensive overhead, and trouble getting funding left Stability floating in troubled waters. And apparently Mostaque sunk the ship himself.

The details:

• Stability CEO, Emad Mostaque, just left the company amid financial turmoil.

• Stability was once raising $100 million at a $1 billion valuation.

• The company has since defaulted on payments to major cloud services, facing severe financial shortfalls.

• Stability AI was left with less than $4 million in the bank by October 2023.

• The company failed to meet numerous ambitious goals set by Mostaque, causing a loss of investor confidence.

• Despite being an open source company, they spent like OpenAI (a lot of money) but without the revenue generating backend.

• Mostaque had a bit of an… ego problem. He had an AI generated Aristotle who announced him as a “modern Prometheus”. 😂 

• Key players within Stability questioned Mostaque’s leadership and resigned.

Why it matters: Talk about a disaster. As CEO, Mostaque was definitely at fault for a lot of Stability’s demise. Which is a shame because open source AI is a good thing. It gives freedom where bigger companies would stifle or censor. But now Stability’s future is uncertain. What a disappointment.

It’s also a warning sign of how these big AI players see themselves and how unstable the industry really is. OpenAI CEO Sam Altman once said he was “shepherding humanity”. Geez. Maybe these people shouldn’t have so much power after all?

📞OpenAI helps businesses create custom models

What happened: OpenAI is expanding their Custom Model program, which is designed to help businesses craft their own AI models to suite their needs. Basically your business trains or creates a custom AI to do repetitive tasks instead of employees.

The details: 

• OpenAI introduced a new part of its program called "assisted fine-tuning" to help companies train AI on specific tasks.

• They're also offering to build entirely custom models for businesses that need a deeper level of customization.

• Big names like SK Telecom are already on board.

• Real applications include tuning AI to chat with customers about telecom and sifting through legal texts.

• OpenAI is looking to rake in over $2B a year with this program. 🤑 

Why this matters: The corporate world is about to adopt AI in a big way. The most repetitive and unimportant tasks will be automated first, things like reading documents and customer service. There’s already cases of AI creating documents only for other AI to then read the documents. But this move paves the way for bigger, smarter AI down the road to do much more. OpenAI is playing big.

😶Meta’s image generator doesn’t like interracial couples

Meta’s image generator is making waves. Turns out it’s impossible to make an interracial couple. Like, at all. Even if you specifically word it multiple ways.

Goes to show that bias is a thing in AI. Whether this comes from the people who programmed and trained it, or just a flaw with the training material… the result means AI can be downright wrong.

Who knows what else AI is biased about?

🏎️These things are too damn expensive!

Source: Midjourney

We’re at your beck and call!

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 ;)