Today in AI

• Today's big story

• News roundup

• AI research changing the world

• This week on the podcast

What’s happening in AI right now

A look at recent AI research

The frontiers of AI research are expanding at a breathtaking pace, with groundbreaking applications emerging across a wide range of scientific disciplines. Recent developments highlight how AI is not just transforming tech-centric fields, but also revolutionizing traditionally data-intensive domains like geosciences. These advancements underscore the growing interdisciplinary nature of AI research and its potential to tackle some of humanity's most pressing challenges.

Earth sciences get an AI boost

The U.S. National Science Foundation (NSF) is making a significant bet on the power of AI to transform our understanding of the planet. The organization is investing over $20 million in 25 projects through its Collaborations in Artificial Intelligence and Geosciences (CAIG) program. This initiative aims to harness cutting-edge AI techniques to address complex challenges in natural hazard forecasting and climate change response.

The funded projects span a diverse range of topics, from earthquake cycles to wildfire prediction and coastal ocean dynamics. By leveraging various AI approaches, including generative AI and causal AI, these initiatives have the potential to revolutionize how we study and interact with our planet.

Challenging core assumptions

As AI continues to make strides in various fields, researchers are also critically examining the fundamental assumptions underlying current AI models. A recent peer-reviewed paper challenges prevailing notions about large language models (LLMs) and their relation to human language. The study argues that many assertions about LLMs' capabilities stem from flawed understandings of language and cognition.

This research calls for a more nuanced understanding of language and highlights the need for rigorous evaluation and auditing of LLMs before deployment. It suggests a paradigm shift may be necessary in AI approaches to language modeling and cognitive replication, encouraging a more interdisciplinary and balanced approach to AI development.

The data dilemma

As AI research advances, the issue of training data availability is becoming increasingly critical. The growing use of robots.txt files to restrict web crawler access is significantly impacting the availability of high-quality training data for generative AI models. This trend is forcing AI companies to reconsider their data acquisition strategies and could potentially alter the future development of artificial intelligence.

The situation highlights the tensions between AI development and content creators' rights, calling for new legal and ethical frameworks for AI training data acquisition. It also raises the possibility of a more fragmented and potentially biased AI training ecosystem, underscoring the need for collaboration between stakeholders to establish fair data usage practices.

News roundup

The top stories in AI today.

FUTURE OF WORK

Jobs are changing — are you?

• ChatGPT Reaches 200M Users as AI Adoption Soars

• New Study Challenges Core Assumptions About AI Language Models

• Oprah’s Special Stirs Controversy

• AI Boosts Productivity Without Mass Job Losses, Study Finds

• AI Engineering Breakthrough Slashes Project Failure Rates

• MIT Expert Warns AI May Not Boost Productivity as Expected

RESEARCH

Breakthroughs, launches, frameworks & more

• AI Tool Predicts Autism in Toddlers with 80% Accuracy

• AI Training Data Shortage Looms as Websites Block Crawlers

• Claude’s AI Boosts Code Editing with New Inpainting Feature

• ChatGPT Unminifies JavaScript Code, Unveiling AI’s Development Potential

• AI Training Data Shortage Looms as Websites Block Crawlers

GADGETS

Computers, phones, wearables & other AI gizmos.

• Canon’s AI Shift Sparks Debate Among Photographers

• Google’s Gemini AI Represents a Significant Leap for Smartphone Capabilities

• Insta360 App Update Boosts AI and Simplifies 360º Video Editing

• AI-Powered Drones Enhance Emergency Response in Switzerland

• How AI-Powered Devices are Accelerating Learning

FINANCE & ECONOMICS

Economics + the movers & shakers on Wall Street.

• NSF Invests $20M in AI to Transform Geosciences Research

• Apple Eyes OpenAI Investment to Boost AI Strategy

• AI Adoption Hits Crossroads as Companies Struggle to Scale

• AI Slashes 1,800 Jobs at Klarna as Fintech Giant Automates

• Meta’s AI Push Drives 34% Surge in Data Center Energy Use

AI research changing the world

The latest breakthroughs and most pivotal papers — broken down in language anyone can understand.

The AI that outperforms humans on college exams

Imagine a world where AI doesn't just respond to queries but actively collaborates on complex projects, speaks multiple languages fluently, and seamlessly integrates text, speech, and visual understanding. This isn't science fiction—it's the reality that Meta's new Llama 3 AI model is ushering in, according to a groundbreaking research report from the tech giant. With its largest variant boasting a staggering 405 billion parameters and the ability to process up to 128,000 tokens of context, Llama 3 is pushing the boundaries of what's possible in artificial intelligence. The report reveals that Llama 3 outperforms humans on college-level exams, demonstrates unprecedented multilingual capabilities across 34 languages, and showcases remarkable prowess in coding and complex reasoning tasks. But the implications go far beyond academic achievements.

This open-source release could democratize access to advanced AI technologies, potentially revolutionizing industries from healthcare to finance, and accelerating scientific discoveries in ways we've only dreamed of. As businesses and researchers begin to harness the power of Llama 3, we're likely to see a surge in AI-driven innovation that could reshape the global workforce and redefine human-AI collaboration. However, this technological leap also raises important questions about job displacement, ethical AI use, and the need for new governance frameworks. As we stand on the brink of this AI revolution, one can't help but wonder: How will society adapt to a world where AI can not only think and communicate at human levels but potentially surpass them in many domains?

The AI tool we’re loving right now

The best way to get AI literate? Try the tools!

This week on the podcast

Can’t get enough of our newsletter? Check out our podcast Future-Proof.

Get more with a Pro account

Paid members get access to discounts on AI tools, expert-written tutorials and deep industry data and leaderboards.