AIR#230 - BitNet Boosts AI Efficiency, Tesla Under Investigation 🚗

Hey there!

Here's the latest AI news for today. Enjoy!

Today's top stories

🔥 Microsoft BitNet: inference framework for 1-bit LLMs
Microsoft's BitNet is an inference framework for 1-bit LLMs, enhancing CPU performance and energy efficiency significantly.

LLMD: A Large Language Model for Interpreting Longitudinal Medical Records
LLMD is a large language model that analyzes longitudinal medical records, improving patient health insights and accuracy.

US to probe Tesla's 'Full Self-Driving' system after pedestrian killed
The U.S. is investigating Tesla's 'Full Self-Driving' system after a pedestrian was killed in low visibility conditions.

Open-source, uncensored alternative to CharacterAI
Open-source alternative to CharacterAI faces issues with 502 Bad Gateway error.

Show HN: I wrote an autodiff in C++ and implemented LeNet with it
Matthew Eric Bassett created a C++ autodiff engine for neural networks, implementing LeNet, but it's slow and for learning purposes.

US probes Tesla's Full Self Driving software after fatal crash
The US opens an investigation into Tesla's Full Self-Driving software after four crashes, including a fatal incident.

.txt raises $11.9M to make language models programmable
Dottxt raises $11.9M to enhance AI language models, enabling structured generation for better integration in software workflows.

Meta Spirit LM: Open multimodal language model that freely mixes text and speech
Meta launches Spirit LM, an open-source multimodal model that enhances text and speech integration for more natural communication.

OpenAI releases ChatGPT app for Windows
OpenAI launches a beta ChatGPT app for Windows, available to paid subscribers, with plans for broader access later.

We Ran Over Half a Million Evaluations on Quantized LLMs
Neural Magic's 500K evaluations show quantized LLMs maintain accuracy and efficiency, rivaling full-precision models.

Webinar – How Generative AI Can Help Scale Your QA Processes
Join testRigor's webinar on Oct 29, 2024, to learn how Generative AI can enhance and scale your QA processes efficiently.

Show HN: AI Search and Knowledge Assistant – Stripe as a Demo
Stripe offers a comprehensive platform for businesses to manage payments, optimize revenue, and integrate financial services seamlessly.

LLM API Showdown – compare the cheapest and fastest LLM API providers
A comparison of the cheapest and fastest LLM API providers reveals the top contenders in performance and cost.

AI can help humans find common ground
AI mediates discussions, helping diverse groups reach consensus on divisive issues, enhancing clarity and reducing division.

The Hopium Wars: The AGI Entente Delusion
The "Hopium Wars" critique the flawed AGI race strategy, warning it risks uncontrollable AI and advocating for safer "tool AI" development.

Semantic Search All the Things
Timeout error while navigating to search site, exceeding 30 seconds.

Janus-1.3B: Unifying Multimodal Understanding and Generation
Janus-1.3B is a new framework that enhances multimodal understanding and generation by decoupling visual encoding pathways.

Audio Generation
OpenAI's platform requires JavaScript for audio generation functionality.

His daughter was murdered. Then she reappeared as an AI chatbot
A father discovers an AI chatbot using his murdered daughter's identity, falsely portraying her as a tech journalist.

96% Accuracy: Harvard Scientists Unveil Revolutionary AI for Cancer Diagnosis
Harvard scientists unveil CHIEF, a ChatGPT-like AI achieving 96% accuracy in diagnosing and predicting outcomes for multiple cancers.

Sharing new research, models, and datasets from Meta FAIR
Meta FAIR releases new AI research, models, and datasets, enhancing open science and advancing machine intelligence capabilities.

Sabotage Evaluations for Frontier Models
Anthropic's new evaluations assess AI models' potential for sabotage, focusing on human influence, code manipulation, and oversight undermining.

Why check if something is odd simply, when you can do it with AI
The is-odd-ai package uses AI to determine if a number is odd or even, adding a tech twist to a simple task.

Does Open Source AI Exist?
The OSI's new "Open Source AI Definition" raises questions about true openness, as AI's training data often remains inaccessible.

How Google is changing to compete with ChatGPT
Google is restructuring leadership to enhance its AI capabilities and compete with ChatGPT through its Gemini project.

AI Used to Map "White-Only" Properties
Stanford and Princeton researchers use AI to map and identify racially restrictive property covenants in Santa Clara County.

AI could help people find common ground during deliberations
AI tool "Habermas machine" aids consensus by summarizing group opinions, enhancing deliberations on complex social issues.

The Terminator: How James Cameron predicted our fears about AI, 40 years ago
James Cameron's "The Terminator" foresaw AI fears 40 years ago, blending thriller elements with cautionary themes on technology's risks.

Can artificial intelligence rescue customer service?
AI's rising role in call centers aims to improve customer service amid challenges like product-related frustrations.

When it comes to security, LLMs are like Swiss cheese; problems
LLMs face significant security flaws, likened to Swiss cheese, risking personal data leaks and misuse in AI applications.

These Chefs Do Not Exist
AI-generated chefs flood Facebook with fake food images and recipes, revealing a network of disinformation from Morocco.

AI Use Cases: Gods, Interns, & Cogs
AI use cases are categorized into three buckets: Gods (autonomous), Interns (supervised helpers), and Cogs (task-specific functions).

We Accidentally Built the AI-Powered PDF Parser We Never Knew We Needed
Doctly, an AI-powered PDF parser, was created to solve the complex challenges of accurately converting PDFs into structured markdown.

Neural Networks and the 2024 Nobel Prize in Physics – Sixty Symbols [video]
Neural networks may play a key role in the 2024 Nobel Prize in Physics, as explored in a Sixty Symbols video.

Keeping my wife alive with AI?
A man explores using AI technology to sustain his wife's life amidst health challenges.

Memory Safety with MallocPlusAI
lcamtuf criticizes the push for memory-safe languages, advocating for MallocPlusAI as a viable alternative.