AI Games are Here! Google Presents ‘Genie’ 🎮

Welcome to another edition of Horizon AI,

Google has recently introduced a new AI model that is set to revolutionize gaming. Meanwhile Mistral unveils its largest language model to date and announces a new partnership.

Let's jump right in!

Read Time: 4.0 min

Here's what's new today in the Horizon AI

• Google DeepMind's AI 'Genie' Can Generate Fully Playable Video Game Worlds 🎮

• Mistral AI Releases New Model to Rival GPT-4 and Its Own Chat Assistant

• AI Research: Study Reveals Users Struggle to Distinguish Between AI-Generated and Human Fake News

• AI Tutorial: Transform drawings into diverse artistic styles

• AI Tools to check out

• The Latest in AI and Tech 💡

AI News

GOOGLE

Google DeepMind's AI 'Genie' Can Generate Fully Playable Video Game Worlds 🎮

Google's DeepMind has introduced Genie, an AI model that can take a text prompt, sketch, or idea and turn it into a virtual world you can interact with and play.

Details:

• Genie was trained on a vast dataset of over 200,000 hours of gameplay videos. It learns from these videos to create controllable virtual worlds without needing specific gameplay instructions.

• Genie generates games with consistent control schemes, even in new AI-generated worlds.

• It can only generate games at a low frame rate for now (1FPS).

• The current focus is 2D platformers; complex 3D worlds remain challenging.

While not yet publicly available, you can find more information and fascinating demos on the official website.

TOGETHER WITH VANTA

eBook: Scaling Your Compliance Program

Do you need to add more security frameworks to your existing compliance program, but don’t know where to start? Scaling your compliance program can feel like you’re proving your security from scratch. It doesn’t have to. 

Vanta’s ultimate guide to scaling your compliance program shares strategies and best practices for adding compliance standards without adding to your workload. 

Learn how to scale, manage, and optimize alongside your business goals.

MISTRAL AI

Mistral AI Releases New Model to Rival GPT-4 and Its Own Chat Assistant

Mistral AI just announced the launch of Mistral Large, a new large language model designed to rival top-tier models like GPT-4 and Claude 2 in reasoning capabilities. Additionally, the company released a beta chatbot service and announced a new partnership.

Details:

• Mistral claims Mistral Large ranks second only to GPT-4 based on the most commonly used benchmarks.

• Its new chat assistant, called Le Chat allows free access to test Mistral Small, Mistral Large, and Mistral Next. The company plans to launch a paid version of Le Chat for enterprise clients soon.

• Mistral announced a new partnership with Microsoft that will give Azure customers access to Mistral AI models.

With these latest moves, Mistral seems to be aiming to position itself as a leader in the AI market.

You can sign up and try out Mistral’s new chatbot on chat.mistral.ai.

Source

AI RESEARCH

Study Reveals Users Struggle to Distinguish Between AI-Generated and Human Fake News

A new study from the University of Lausanne and the Munich Center for Machine Learning reveals that people can't tell the difference between fake news crafted by humans and AI.

Details:

• Researchers had nearly 1,000 participants rate fake news stories about COVID-19. Some were written by humans and others used GPT-4.

• 12% of people were willing to share these fake stories on social media, whether human or AI-generated. The researchers found no statistically significant difference in people's willingness to share human-generated fake news compared to AI-generated fake news.

• Certain demographics like older social media users and those with less education are more vulnerable

The study recommends implementing new labels to differentiate real and fake content, along with educating people about AI-generated fake news risks.

Source

AI Tutorial

Transform drawings into diverse artistic styles

1. Go to Drawn to Style GPT

2. Upload your drawing and enter your prompt

transform this sketch into [specific style]

Here are some style ideas:

1. Anime Style

2. 1950s Comic Illustration Style

3. Neon Style

4. Photorealistic Style

5. Pixar 3D Style

6. Traditional Animation Style

7. Water Color Painting

Example:

AI Tools to check out

✍🏻 Cartoonifier: Create cartoon characters of yourself with AI.

✅ PoplarML: Enables the deployment of production-ready, scalable ML systems with minimal engineering effort.

🤖 Dart: Project management tool that uses AI to assist with roadmap planning, report generation, and more.

🖼 Synthesys X: Generates new images based on the objects and patterns detected in a given image, considering their similarities and relevance.

👨‍💻 Danswer: AI-powered search engine across your company's docs, apps, and people.

The latest in AI and Tech

Google aims to relaunch Gemini AI image tool in a few weeks

Google plans to relaunch its AI image generator, Gemini, in a few weeks after pausing it last week. Users discovered historical inaccuracies and questionable images, which went viral. The tool was not working as intended, said DeepMind's CEO.

No More Apps? Deutsche Telekom's Wild New AI Phone

Deutsche Telekom recently showcased a novel smartphone concept that forgoes apps and instead uses AI to handle users' needs. Developed in collaboration with Qualcomm and Brain, the app-free interface would allow the device to respond to prompts, provide tailored recommendations for purchases or travel, and share media with contacts.

Meta is preparing for the EU’s 2024 Parliament Elections

In preparation for pivotal European Parliamentary elections, Meta is rolling out new initiatives aimed at mitigating the impact of misinformation on its platforms. Measures include setting up an EU-specific Election Operations Center, augmenting its fact-checking network, and developing tools to detect and label AI-generated content.

Jensen Huang says kids shouldn't learn to code

Speaking at the recent World Government Summit in Dubai, Nvidia CEO Jensen Huang made the bold prediction that advances in AI will soon eliminate the need for humans to learn programming languages. Learning to code should no longer be a priority for those looking to enter the tech sector.

Alphabet's AI problems: Stock falls 4%

Alphabet shares plunged 4.4% on Monday following last week’s pause of Gemini’s image generation feature that drew criticism over inaccurate historical depictions of race. Alphabet's market value decreased by over $70 billion. Analysts suggest that these recent controversies may contribute to the perception of Google as 'an unreliable source for AI' and potentially open doors for competitors.

That’s a wrap!

Thanks for sticking with us to the end! Let’s stay connected on LinkedIn and Twitter.

Not subscribed yet? Sign up here and send it to a colleague or friend!

See you in our next edition!

Gina 👩🏻‍💻