Good morning. It’s Monday, May 13th.

Did you know: This week in 1997, The IBM computer and artificial intelligence Deep Blue defeated reigning chess champion Garry Kasparov?

In today’s email:

• OpenAI’s Live Updates

• Google’s Announcements for Tuesday

• Can AI Date For You?

• 5 New AI Tools

• Latest AI Research Papers

• AI Creates Comics

You read. We listen. Let us know what you think by replying to this email.

Watch the OpenAI Announcement Live here at 9:30am PT

ChatGPT To Get Major Upgrades, Including New Voice Features, Agents And A New Interface

OpenAI's upcoming tweaks to ChatGPT are generating a buzz and a promise on a wave of enhancements including new voice capabilities, sophisticated agents, and a revamped user interface.

Rumors of GPT-5 or a revolutionary search engine might be taking a backseat as they appear to be optimizing the user experience and strengthening ChatGPT's role as a capable AI assistant. This pragmatic strategy suggests a focus on iterative improvements rather than a moonshot for entirely new functionalities.

The forthcoming assistant is anticipated to excel in speech and image recognition, as well as logical reasoning, all integrated into a more efficient and cost-effective model compared to its predecessors. However, OpenAI CEO Sam Altman appears to be gazing further into the future, envisioning a world where these AI assistants transcend their utilitarian roles and evolve into something far more nuanced – sophisticated companions reminiscent of those portrayed in science fiction films like "Her."

Potential improvements for ChatGPT, as speculated by developer Tibor Blaho, encompass a gamut of enhancements including interface redesign, advanced voice mode, custom GPTs, cloud service connectors, enhanced search capabilities, and augmented chat-sharing options, promising a more immersive and intuitive user experience. Read more.

Google I/O 2024 Will Be All About AI Again

Google is gearing up for its annual developer extravaganza, Google I/O, set to kick off with an AI-centric keynote on May 14th at 1PM ET. With the spotlight firmly on artificial intelligence, Google aims to showcase its latest strides in AI technology, building upon the foundation laid by Gemini, its formidable counterpart to OpenAI's ChatGPT. Following a year of intensive testing and refinement across Search, Google Maps, and Android, the tech titan is poised to unveil a plethora of cutting-edge features, promising a captivating exploration into the future of AI integration.

Anticipation is high for Google's fusion of Search and generative AI, expected to unlock innovative functionalities such as AI-driven conversation practice and image generation for immersive shopping experiences. Moreover, rumors swirl of a game-changing digital assistant dubbed "Pixie," harnessing Gemini's prowess to deliver a deeply personalized user experience. As Google elevates the smartphone into an AI powerhouse, competitors in the burgeoning AI wearable market may find themselves outmaneuvered, signaling a seismic shift in the AI landscape. Read more.

Bumble's Founder Says AI Could Date 'Hundreds' Of People For You

Whitney Wolfe Herd, founder and executive chair of Bumble, envisions a future where dating is transformed through AI, with chatbots streamlining the matching process. She foresees a scenario where an AI dating concierge interacts with other concierges on behalf of users, sparing them from engaging with hundreds of potential matches. This transformation aligns with Bumble's commitment to fostering safer and more equitable digital platforms, emphasizing empowerment, especially for women.

As AI increasingly permeates dating apps, Bumble aims to integrate AI functionalities to coach users on dating strategies and self-presentation. Read more.

Etcetera: Stories you may have missed:

> Nvidia launches quantum computer centers with CUDA-Q platform 

> AI GPU bottleneck has eased, but now power will constrain AI growth warns Zuckerberg

> Apple’s New ChatGPT Deal—Here’s What It Means For Your iPhone

> DeepMind is experimenting with a nearly indestructible robot hand

> Deepmind proves robust AI adaptation requires learning causal models under the hood

> Japan team uses Fugaku supercomputer to develop language model for AI  

> Exploration-focused training lets robotics AI immediately handle new tasks

> OpenAI thinks it knows what media is "high quality," and that's a problem  

Pictographic is an AI-generated illustration library. Explore 40K+ images and SVGs or create custom illustrations instantly. Intuitive platform for color adjustment. No payment needed.

Smartrazor simplifies YouTube video editing for talking head creators. Automates A-roll editing, with B-roll support planned. Enhance workflow efficiency.

AI Logo Design by Stylar offers an all-in-one solution. Elevate logo design with 3D text effects, easily adding depth and personality.

Relume is an AI-driven website builder for rapid creation of sitemaps, wireframes, and UI style guides. Streamline workflow, reduce project timelines, enhance collaboration.

CallFluent AI is a cloud-based software that automates inbound and outbound calls with AI-powered voice agents. Enhances operational efficiency, customer satisfaction, and revenue opportunities.

📄 Characterizing the Accuracy - Efficiency Trade-off of Low-rank Decomposition in Language Models

📄 A Survey on RAG Meets LLMs: Towards Retrieval-Augmented Large Language Models  ​​

📄 CatLIP: CLIP-level Visual Recognition Accuracy with 2.7x Faster Pre-training on Web-scale Image-Text Data  

📄 PuLID: Pure and Lightning ID Customization via Contrastive Alignment

📄 BASS: Batched Attention-optimized Speculative Sampling 

Thank you for reading today’s edition.

Your feedback is valuable. Respond to this email and tell us how you think we could add more value to this newsletter.

Interested in reaching smart readers like you? To become an AI Breakfast sponsor, apply here.