Meta Used Your Facebook and Instagram Posts to Create AI Assistant

If the thought of an AI model trained on Facebook and Instagram posts frightens you, boy, do we have news for you.

Meta wowed crowds at the company's annual Connect conference last week after CEO Mark Zuckerberg announced the launch of Meta AI, a virtual assistant partly trained on publicly available Facebook and Instagram posts. We're sure someone, somewhere thought this was a good idea, and in practice, it might be, but the idea that our collective posts served as the basis of training yet another LLM should ring alarm bells.

Meta's top policy executive clarified that a vast majority of the data used to train Meta AI was publicly available, and that the company shunned private posts and texts, presumably, so users don't become too uncomfortable with the idea of having their data trails serve as the basis of an AI that ultimately benefits one of the most evil corporations in the world.

As if the details on exactly how Meta AI was trained weren't grim enough, the company's President of Global Affairs Nick Clegg told Reuters that Meta believed any copyrighted material used in training the assistant, which was built using a custom version of Meta's Llama 2 large language model, fell under the existing fair use doctrine. Of course, the company has lawyers at the ready if creators feel otherwise because Clegg then added that Meta expects a "fair amount of litigation" on the use of copyrighted materials and suspected all of this was "going to play out in litigation."

When all is said and done, it seems Meta too is interested in the AI pie to remain relevant, having made no significant mention of augmented and virtual reality at the Connect conference (Zuckerberg's obsession up until recently).

Facebook ranked #1 on HackerNoon's Tech Company Rankingsthis week. Instagram was in the #7 spot.

After Stealing the Internet, Is Amazon Eyeing AI Next? 👁️

Not to be left behind by some of its peers, Jeff Bezos' Amazon has decided to invest up to $4 billion in a San Francisco-based AI startup called Anthropic to rival the likes of Microsoft's OpenAI, the creator of the hit ChatGPT.

The ecommerce giant has made an initial investment of $1.25 billion with an option to pay another $2.75 billion that can be triggered by either side. The total investment is valued at an eye-popping $4 billion.

According to media reports, Anthropic is the developer of a generative AI model called Claude that is designed to rival OpenAI's ChatGPT. In Amazon's case, the company plans to use the capabilities of Claude to enhance customer experiences, which means you can expect it to come to an Alexa near you.

Other media reports indicated that Amazon's employees and cloud customers will gain early access to technology from Anthropic as part of the deal.

All in all, it looks like the AI space is shaping up to be a hot bed for mergers and acquisitions. As mature companies snap up promising companies in the space, one has to wonder: after stealing the internet, are tech giants eyeing AI next?

Amazon ranked #14 on HackerNoon's Tech Company Rankings.

\In Other News.. 📰*

• AMD CEO Lisa Su on the AI revolution and competing with Nvidia — via The Verge.

• Elon Musk's X Corp in another legal fight over unpaid rent, this time as plaintiff — via Reuters.

• This is Financial Advice — via YouTube.

• Google launches $399 Chromebook Plus category — via TechCrunch.

• Atari 2600+ sees its future in retro gaming — via CNN.

• How AI works, in plain English: Three great reads — via Axios.

• Apple will issue a software update to address iPhone 15 overheating complaints — via CNBC.

And that's a wrap! Don't forget to share this newsletter with your family and friends! See y'all next week. PEACE! ☮️

— Sheharyar Khan, Editor, Business Tech @ HackerNoon

*All rankings are current as of Monday. To see how the rankings have changed, please visit the Tech Company Rankings page.