Issue 56 11th June 2024

Highlights

Apple brings ChatGPT to Siri as it debuts ‘Apple Intelligence’ at WWDC 2024

New features and deal with OpenAI presented at conference marks change in focus for tech giant, which is under pressure to catch up with rival firms’ AI push Tim Cook, the Apple CEO, announced a series of generative artificial intelligence products and services on Monday during his keynote speech at the company’s annual developer conference, WWDC, including “Apple Intelligence” and a deal with ChatGPT-maker OpenAI. The new tools mark a major shift toward AI for Apple, which has seen slowing global sales over the past year and integrated fewer AI features into its consumer-facing products than competitors. Continue reading...

This Week in AI: Ex-OpenAI staff call for safety and transparency

Hiya, folks, and welcome to TechCrunch’s inaugural AI newsletter. It’s truly a thrill to type those words — this one’s been long in the making, and we’re excited to finally share it with you. With the launch of TC’s AI newsletter, we’re sunsetting This Week in AI, the semiregular column previously known as Perceptron.

Satya Nadella has made Microsoft 10 times more valuable in his decade as CEO. Can he stay ahead in the AI age?

Ten years in as CEO, Nadella has turned Microsoft into the world’s most valuable company and one of the top players in AI.

Paper of the week

What happens if you remove matrix multiplication from your LLM?

That LLMs are quite energy hungry and slower than we'd like doesn't come as a surprise. But what can you do? Well, you could, for example, remove all the multiplications to save 61% memory and reduce power usage to 13 watt instead of multiple kilowatts.

This paper demonstrates with code how LLMs can be made quite a bit more efficient when we take a radically different approach.

Video

Copilot L33t Sp34k | MSRC Bug Bounty

Sarah talks to Lynn Miyashita and Arnav Garg from the Microsoft Security Response Center (MSRC) about the new AI bug bounty and how Microsoft investigates AI bugs reported via the MSRC.

AI Show: On Demand | Boosting Phi-3’s Language Capabilities with Azure AI Translator

In this episode of The AI Show, we explore how Azure AI Translator enhances Phi-3's multilingual translation quality. Discover the transformative impact of integrating advanced algorithms for low-resource languages, and how this synergy bridges communication gaps with remarkable precision.

Articles

How I Built an LLM-Based Game from Scratch

A few months ago, I was debating with a friend about the use of Large Language Models (LLMs) in game design. My friend argued that LLMs were too unpredictable to be used consistently in games and should not be used for “live mechanics,” while I believed they could propose innovative experiences if controlled through the right framework.

Introducing GraphRAG with LangChain and Neo4j

GraphRAG is an approach that relies on storing the knowledge base into Graph databases, making the retrieval phase more efficient and henceforth leading to a more relevant contex to then let the LLM generate the best answer.

How to spot a deepfake: the maker of a detection tool shares the key giveaways

Siwei Lyu of the DeepFake-o-meter explains how to tell when photos, videos and audio aren’t real You – a human, presumably – are a crucial part of detecting whether a photo or video is made by artificial intelligence. There are detection tools, made both commercially and in research labs, that can help. To use these deepfake detectors, you upload or link a piece of media that you suspect could be fake, and the detector will give a percent likelihood that it was AI-generated

Accidental prompt injection against RAG applications

The question here was “What is the meaning of life?”—my LLM documentation tries to be comprehensive but doesn’t go as far as tackling grand philosophy!

Building LLM Apps: A Clear Step-By-Step Guide

Comprehensive Steps for Building LLM-Native Apps: From Initial Idea to Experimentation, Evaluation, and Productization Through this experience, I developed a battle-tested method for creating innovative solutions (shaped by insights from the LLM.org.il community), which I’ll share in this article.

Thoughts on the WWDC 2024 keynote on Apple Intelligence

Today's WWDC keynote finally revealed Apple's new set of AI features. The AI section (Apple are calling it Apple Intelligence) started over an hour into the keynote - this link jumps straight to that point in the archived YouTube livestream.

INT4 Decoding GQA CUDA Optimizations for LLM Inference

An efficient decoding Grouped-Query Attention with low-precision KV cache

A technique for more effective multipurpose robots

With generative AI models, researchers combined robotics data from different sources to help robots learn better.

How I Use ChatGPT As A Data Scientist

How ChatGPT improved my productivity as a data scientist.

Upcoming events

Join the Global DevOps Experience on June 15, 2024!

The Global DevOps Experience is a unique, global event where the worlds of DevOps, Platform Engineering, Developer Experience (DevEx), and AI come together. This one-day event will be hosted at many venues all over the world and is more than an event; it is an experience! Participants will be part of a team in a fictitious company, Globoticket, and help them overcome some difficult and relevant challenges in the fields of Software Development with AI, Platform Engineering, and DevOps.