This Week in AI

Amazon Cloud CEO Predicts Future Where Most Software Engineers Don’t Code—AI Does It

Matt Garman, CEO of Amazon Web Services, told employees that in about two years, “it’s possible that most developers are not coding.” He expects AI to handle most of the coding, allowing developers to spend more time understanding customer needs and guiding AI to create software. Business Insider obtained his remarks after they were leaked. Perhaps Garman was thinking about Devin, the AI software engineer created by Cognition AI, that is currently a leader in the space.

Epic Touts New AI Applications to Improve Care and Speed Charting

Medical health records giant Epic is bringing new generative AI features to its platforms to streamline charting, ease clinician workloads, and improve patient experiences. The system augments doctor-patient communication by automatically drafting responses, saving doctors up to half a minute a message. Across 150 healthcare systems, it’s drafting one million messages PER MONTH. That’s a lot of clinician minutes saved, especially when 40%-60% of clinicians report burnout due to staff shortages. In testing, patients prefer so-called augmented responses, remarking they seem more human—Perhaps more of a commentary on clinician overload than anything else. The Epic system can auto-chart patient-doctor conversations, enabling the clinician to focus on their patient rather than note-taking. Epic says they have another 100 generative AI projects in flight.

New AI Algorithm Can Predict Tipping Points for Future Disasters

Researchers have published details of a new AI model that predicts the onset of catastrophic tipping points in complex systems, a task that has proven very difficult, until now. Tipping points can occur in any complex system, for example, financial markets, ecological systems, climate change, or urban traffic networks. Dramatic transformations in complex systems are poorly understood, and scientists’ predictions often rely on oversimplified models, making predictions shaky. The new AI approach improves prediction by modeling complex interactions and crunching far more data than a human mind can comprehend.

Quick Hits

• OpenAI and Condé Nast Deal - The publishers of The New Yorker, Vogue, Bon Appétit, Wired, and Vanity Fair signed a multiyear licensing deal related to ChatGPT and OpenAI’s new Google competitor, SearchGPT. 🔍

• U.S. Government Bans AI-generated Reviews - The Federal Trade Commission plans to crack down on fake reviews of all kinds, including AI-generated ones.

• AI Could Turn Anyone into an Expert Animal Tracker - An AI system named WildTrack can successfully identify 17 species from their footprints and will soon identify individual animals. 🐾

• Alien: Romulus Brings Back Dead Actor with AI - Ian Holm died in 2020, but thanks to AI, he plays a role in the latest Alien franchise movie. Controversy and outrage have ensued.

• Trump Upsets Swifties - Furious Taylor Swift fans aren’t happy after Trump shared a fake endorsement by the singer featuring AI-generated images.

Video: AI Music Made by a Musician 🎶

In this thoughtful and thought-provoking video, experienced musician Jonny Keely talks about using Udio to help him make a song and the huge questions it poses for the future of music. The song he creates with AI is impressive, and Keely draws a sharp contrast by singing a version himself. You’ll be surprised by which he prefers. It’s worth watching if only to understand just how good AI music generators are today. (10m)

AI Tech and Innovation

Eric Schmidt’s AI Prophecy: The Next Two Years Will Shock You

Former Google CEO and now AI mega investor Eric Schmidt recently spoke to students at Standford University and dropped some big ideas on the audience. Schmidt believes the potent combination of enormous context windows, agentic AI, and text-to-action models will lead to new AI capabilities that accelerate innovation exponentially within just a few years. He also predicts the scale of AI investments will skyrocket. At Schmidt's request, online copies of the video interview keep being deleted due to his comments about Google’s remote work policy, but here’s a link to try. 🔮

These ‘Living Computers’ Are Made From Human Neurons - Rent One for $500 a Month

Transistors power the modern world but suck enormous amounts of power. By comparison, the neurons in our brain are incredibly energy efficient. The human brain consumes only about 25W while working flat out. Our brains prove that high-level intelligence doesn’t have to consume megawatts of power. Scientists are now building computers using synthetic biology to grow clusters of ‘organoids,’ lab-grown 3D structures that resemble human tissue. Swiss company FinalSpark has built a “Neuroplatform” that scientists can rent over the net for $500 a month. Their goal is to create AI that consumes 100,000 times less energy than traditional methods. 🧠

AMD Is Looking to Catch Nvidia with ZT Systems Deal

All chipmakers are eyeing Nvidia’s fat margins and profits. One reason Nvidia has such a strong leadership position is that it doesn’t just sell chips; it sells complete AI solutions, including chips, boards, networking, server building blocks, and a comprehensive software stack. AMD picked up server builder ZT Systems for a cool $4.9 billion, so it has “a more complete offering” and an edge in designing systems. Chip designers constantly need feedback from systems and model builders as they create their next-generation products.

Benchmarks of Almost Every Multimodal LLM

If you’re in IT and trying to decide which large language model to use for your next project, this paper might help you choose. It comprehensively benchmarks 180 leading models and evaluates their strengths and weaknesses. 📊

AI Insights

New ‘AI Scientist” Can Write Science Papers Without Human Input. That’s a Problem.

One of the great promises of AI, particularly artificial general intelligence, is the ability to automate or semi-automate scientific research, thereby accelerating discovery and yielding incredible breakthroughs in medicine, material science, and energy. This has been a significant focus of AI research labs like DeepMind with their AlphaFold and GNoME platforms. Now Sakana AI has announced an ‘AI scientist’ designed to automate scientific discovery and write research papers. Some have described its output as “endless scientific slop” and argued that AI tools should support scientists, not seek to replace them. 🔬

Here’s Why the Debate Over Open Source Matters For Us Humans

A debate is raging in AI circles about the benefits and risks of an open-source approach to AI. Open-source projects gave us many of the internet’s underpinnings. Linux, WordPress, and Apache rule the web. Kubernetes and Docker make the cloud possible. Facebook, Instagram, Reddit, YouTube, and Spotify rely on open-source relational databases MySQL and PostgreSQL. Many AI startups build on top of open-source AI models from Meta, Mistral, Salesforce, and other companies to add value and fine-tune them for specific applications. Others argue that as AI models become more powerful, open-source AI could enable bad actors and amplify AI safety risks.

Ranked: Countries That Use ChatGPT the Most

Which country do you think has the highest consumer use of ChatGPT as a percentage of the population? The United States isn’t even in the top ten. In the leading country, 45% of consumers are using ChatGPT.

Toolkit for the Future

Reclaim.AI: The AI assistant for your calendar

Maintain healthy calendar habits, improve your productivity, optimize cross-team meetings, boost collaboration, and improve work-life balance. Reclaim schedules 1:1s and defends your calendar so you can focus. Syncs multiple calendars and integrates with Google Calendar, Zoom, Slack, and HubSpot.

SupportBench: Elevate customer satisfaction with #1 rated help desk

Take customer support to the next level with powerful AI features including chatbots, workflow automation, emotion insights, knowledge bases, data-drive insights, real-time reporting, and empathy coaching to help your support teams increase productivity and customer satisfaction.

Firstbase: The easy way to start a new company

Incorporate your company, access one-click growth tools, stay compliant, and manage everything your business needs — all online, from anywhere. Launch your U.S. business in minutes with no paperwork or legal headaches.

Oyster: Find and hire international talent with a simple cross-border HR solution

Need to expand internationally, but not sure where to start? Oyster makes it easy to find, hire, and retain local talent. They handle the details (country-specific labor laws, international tax laws, compliance, and global payroll) so you can focus on finding the right talent and ramping their impact.

Manifest: Boost website sales with this chatbot

Build intelligent personal shopping experiences by adding the Manifest AI chatbot to your site. Help shoppers find what they need, faster. Double add-to-cart and conversion rates and get 25% higher AOV. Easy Shopify and help desk integration. Free 14-day trials.

ThirstySprout: Hire top-tier tech talent, fast

Hire top talent with industry-specific expertise to build world-class engineering teams and solve complex business problems. AI-assisted sourcing, vetting, and hiring delivers the right talent, at the right time, globally!