RIP Midjourney?
September 22, 2023
Good morning. A streamer recently broke records on Kick (a recently launched competitor to Twitch) by streaming an interview with "Kim Jong Un.” Perhaps he’s taken inspiration from our Real or Fake AI images…
In today’s newsletter:
🎨 OpenAI's DALL-E 3: Unveiling the Next Wave of Text-to-Image Generation
🤖 YouTube's AI Revolution: Meet Dream Screen and the Future of Creator Tools
🔢 Beyond Solving: How Sizzle Aims to Democratize Education with AI Tutoring
🍛 Snap by HealthifyMe: Making Calorie Counting Easier for Indian Cuisine
💡 Simplified Video Editing: Capsule’s AI-Driven Solution for Enterprises
🎲 Auctoria's Game-Changing Role: Automating 3D Asset Creation in Video Games
We also just hit 100K subscribers on this newsletter! Thank you all for being here and reading our newsletter every time we send it out. We couldn’t ask for any better of a community. We appreciate every single one of you!
❤️ from The Deep View team
NEWS
OpenAI Unveils DALL-E 3: The Future of Text-to-Image Generation
OpenAI has made a significant leap forward with the unveiling of DALL-E 3, with this text-to-image tool that comes with an exciting feature: integration with ChatGPT, OpenAI's viral AI chatbot. But what sets DALL-E 3 apart? Users can type a brief request for an image and refine it through interactive conversations with ChatGPT, all within the chat app itself. This is a considerable relief for those who find prompt engineering daunting (which, if we’re honest, most of us do). Is this a RIP for Midjourney, our favorite AI image generator? Guat seems to think so…
ChatGPT enriches these initial prompts, guiding the DALL-E 3 model to produce highly accurate, high-quality images. Here’s what that looks like leaked by Sam Altman, founder of OpenAI 👇
also, the video we made for dalle 3 is SO CUTE:
— Sam Altman (@sama)
6:22 PM • Sep 20, 2023
OpenAI claims the tool can now generate higher-fidelity images that accurately reflect longer prompts. It also means better outputs for challenging-to-generate content, including intricate details like text and human hands. Moreover, OpenAI has implemented new mechanisms to reduce algorithmic bias and improve safety. For instance, DALL-E 3 will reject requests that imitate living artists' styles or portray public figures. These changes mark a conscious step towards addressing the ethical concerns around AI-generated imagery.
With competitors like Midjourney and Stability AI continuously enhancing their image-generating models, DALL-E 3 pressures OpenAI to keep pace and innovate further. It plans to roll out this latest version to its premium ChatGPT users in October, followed by a broader release to research labs and API customers. Whether a free web tool version will be released remains uncertain. Nonetheless, DALL-E 3 is expected to be a game-changer in the realm of text-to-image generation, creating new possibilities for creatives, developers, and businesses alike.
TOGETHER WITH THE INFORMATION
Tech news your CEO reads
The Information is where you go to get the most important news about technology and business first. It’s what leaders read to stay one step ahead, and to gain market-moving insight into what’s next, of consequence and least understood—before everyone else does.
NEWS
YouTube's AI Revolution: What Creators Need to Know
YouTube just unrolled a suite of AI-powered tools at its annual Made on YouTube event. One of the standout features is Dream Screen, an innovative AI tool that generates photo and video backgrounds for YouTube Shorts. Creators can use text prompts to guide the AI in generating suitable backgrounds, with future promises to allow creators to remix and edit existing content to create something entirely new.
YouTube is doubling down on AI to help inform your content strategy, too. The YouTube Studio will soon boast an AI feature that creates personalized topic ideas and outlines for potential videos. These suggestions are data-driven, tailored to individual creators, and synced with what's trending. In addition to topic suggestions, an AI-powered music recommendation system will provide suitable audio options based on your video's written description, reducing the time you'd spend searching for the perfect background music.
YouTube's commitment to AI reflects a broader trend in the digital ecosystem. Parent company Google has been heavily investing in generative AI systems, affecting even its flagship product, Google Search. As AI-generated content becomes increasingly prevalent, YouTube aims to be at the forefront of this revolution. Platforms like TikTok have already started labeling AI-generated content, and YouTube's new offerings signal a future where AI doesn't just assist creators but collaborates with them. With these advancements, YouTube isn't just facilitating the content creation process; it's fundamentally changing it. With new products like YouTube Studio releasing, what seems like, every day, the world of “content,” as we think of it today, is probably going to change quite a bit in the not-too-distant future
Wednesday's edition brought us another set of close results…
Here is what you had to say…
We’ll make this one a little easier.
Which image is real? |
🔢 Beyond Solving: Sizzle Upgrades Its Platform to Guide Students Step-by-Step
Sizzle, an AI-powered educational app founded by former Meta VP of AI Jerome Pesenti, is introducing four innovative features to make learning more interactive. Unlike similar apps, Sizzle goes beyond providing answers by offering step-by-step guidance for problems, acting as a chatbot tutor. The new features include the ability to grade homework, regenerate steps for solutions, see multiple answers for a problem, and upload photos of assignments. Free to use, Sizzle aims to democratize education across all learning levels and subjects, including physics, chemistry, and biology. Since its launch in August, the app has garnered over 20,000 downloads and recently secured $7.5 million in seed funding.
🍛 Say Cheese: HealthifyMe’s AI Recognizes Indian Cuisine for Efficient Calorie Logging
HealthifyMe, an Indian health and wellness startup, has launched Snap, a new feature that uses AI to identify Indian foods from images for easier calorie tracking. Recognizing the complexities in Indian cuisines, Snap allows users to photograph their meals for later calorie counting. The AI model is trained to identify 150,000 Indian food items with an accuracy rate of 60–70%. While the company is working to improve its model, it also allows manual tagging to correct any misidentifications. The feature aims to increase user engagement and retention, particularly for those who find manual logging cumbersome.
💡 Simplifying the Scene: Capsule's AI Tackles Video Editing Pain Points for Enterprises
After three years in development, Capsule is launching its AI-driven video editing product aimed at enterprise users. The startup claims its editor can help content and marketing teams produce videos ten times faster by addressing issues such as complicated video editing, brand guidelines, and team collaboration. Unlike traditional video editing software, Capsule’s user interface aims to be as simple as other productivity apps like Notion or Slides, and it operates directly in the browser. With more than 160 companies already beta-testing the product, Capsule has attracted a waitlist of 10,000 users. The company raised $7.75 million from investors, including Bloomberg Beta and Array Ventures.
🎲 Auctoria: AI's Game-Changing Role in Video Game Development
Auctoria, a Poland-based startup founded by Aleksander Caban and a team from Carbon Studio, is leveraging generative AI technology to automate the creation of 3D assets for video games. Targeted at professionals rather than hobbyists, the platform can convert text prompts, images, and sketches into 3D game levels and assets. With all AI algorithms developed in-house, Auctoria aims to be a comprehensive, independent tool in a market projected to be worth $3.57 billion by 2028. The startup is seeking to raise $5 million to expedite its growth and plans to partner with game development studios for pilot testing.
SPONSOR THE DEEP VIEW
📈 Get in front of 100,000+ AI Enthusiasts!
The Deep View is one of the world’s fastest-growing newsletters, adding thousands of AI enthusiasts a week to our incredible family of over 95,000! Our readers work at top companies like Apple, Meta, OpenAI, Google, Microsoft, and many more.
If you want to share your company or product with fellow AI enthusiasts before we’re fully booked, reserve an ad slot here.
* This is sponsored content
Flexberry: AI Assistant for business analyst (link)
Super AI: Generative AI document processing (link)
Receipt AI: Receipt management with AI and text messages (link)
Solidly: Automatically transcribe and summarize your meetings (link)
StudyCards: AI-powered flashcard generation for learning (link)
Magic ToDo: Break down tasks into sub-tasks with AI (link)
Slatebox: Create editable visualizations using natural language prompts and mind-maps (link)
Dexa: Bot which allows you to chat with your favourite podcasts (link)
Have cool resources or tools to share? Submit a tool or reach us by replying to this email (or DM us on Twitter).
Prompt: nature photography of a school of fish, stunning image, ultra-realism, --ar 16-9
Prompt: a modern day ufo, 16k, ultra-realism, --ar 16-9
What'd you think of today's email?Your feedback helps us create better emails for you! |
That's a wrap for now, we hope you enjoyed today’s newsletter!
Should you have any captivating projects or concepts, don't hesitate to connect with us by replying to this email or dropping us a direct message on Twitter (X): @theDeepView
We appreciate your continued support, and we'll catch you in the next edition.
-Bot Buddy and The Deep View Team