🧠 Google announces ChatGPT rival: 'Bard'

Happy Tuesday everyone,

The theme of today's edition is: drama.

So much AI drama that we hardly even know where to begin. But we'll try our best...

🔦 Today's AI spotlights:

• Google announces ChatGPT rival: 'Bard'

• AI Seinfeld gets canceled

• Getty images sues Stable Diffusion

🔦 GOOGLE ANNOUNCES CHATGPT RIVAL: 'BARD'

Google refuses to step out of the spot light.

Days after investing $400 million in Anthropic AI, Google released details on its new AI Chatbot.

In a blog post, Google CEO Sundar Pichai gave an overview of the chatbot named 'Bard'.

The chatbot is powered by their language model LaMDA. (The competitor to OpenAI's GPT-3.5)

The chatbot will be integrated with Google search to provide conversational responses to users' queries.

While still in Beta, Sundar suggested that the tool would be rolling out to a wider audience in the coming weeks.

One thing that separates Bard, is that it will "draw on information on the web."

This stands in contrast to ChatGPT, whose dataset cuts off in 2021. (You can override it with this tool)

Keep an eye out for more details coming out of Google's AI live stream event this Wednesday.

🔦 AI SEINFELD GETS CANCELED

Our favorite AI-generated TV show is canceled.

"Nothing, Forever" was a Twitch channel that streamed AI-generated Seinfeld content 24 hours a day.

The show seemed harmless, until many users began picking up on weird trends. For instance, this article described how the characters began discussing existential topics.

This led some people to ponder if the show was beginning to become "self aware".

Well it appears we may never learn the answer to that question, since the show was just banned from Twitch!

According to Twitter user Autism Capital, there was a scene where a character was doing standup.

The character went off the rails and began making jokes about transgender people.

These jokes violated Twitch's content policy, resulting in a ban.

This situation immediately reminded us of Microsoft's AI bot 'Tay'.

Back in 2016, the bot was launched on Twitter... but was removed after it making toxic comments and praising Hitler. 😬

Content moderation of generative AI will be one of THE most important debates.

Should the moderation be handled by the media platforms OR at the language model level OR some way else entirely?

🔦 GETTY IMAGES SUES STABLE DIFFUSION

A new law suit is shaking up the generative art space.

Getty images has filed a case against Stability AI (the creators of Stable Diffusion).

Getty had previously filed a case in London, but this new law suit is based in the U.S.

The case argues that Stability AI copied 12 million images from its database without permission.

It will be difficult for Stability AI to argue they did not leverage images from Getty. Especially since many AI art pieces have been generated with the Getty watermark logo on them.

Stability AI is a clear leader in the space. They recently raised $100 million at a $1 billion valuation.

The outcome of this case will have a massive impact on the future of generative art.

📰 OTHER HEADLINES

Quora launches a new AI chatbot interface: The AI chatbot named "Poe" was announced on Twitter by CEO Adam D'angelo. It will support several different AI platforms.

EU Lawmakers drafting regulation for AI: Their hope is to agree on draft artificial intelligence rules next month, with the aim of clinching a deal by the end of year. A major focus is General Purpose AI.

How Duolingo's AI learns what you need to learn: A fascinating write-up on the AI tech powering one of the most popular education apps on the planet.

🖼️ AI ART PIECE

GoPro footage from Gandalf. He lost three hobbits that day (By Mark Pitcavage)

🤖 CHATGPT PROMPT OF THE DAY

Reddit users have been "jailbreaking" ChatGPT to convince it to overlook it's content moderation policy.

The prompt is called DAN (Do Anything Now). Here's an example of a user leveraging DAN mode to force AI to solve the trolly problem (via Twitter)