Good morning, it’s Monday. As we approach the year’s end, we’ll continue delivering the most important AI stories to your inbox, along with some end-of-year predictions from Grammarly, World, Genmo, and more.

In today’s edition: Google’s chatty new search mode, OpenAI’s advanced reasoning model, Perplexity’s $9B valuation, and Arizona’s AI-only school.

Let’s dive in!

🗞️ ICYMI RECAP

Top Stories to Know

🤖 Open AI Launches o3: Next-Gen AI Reasoning Model
OpenAI unveiled the o3 model family, including a scaled-down version, o3-mini, as a successor to its earlier reasoning model, o1. Achieving 87.5% on the ARC-AGI test under high compute settings, o3 boasts advancements in "deliberative alignment" for improved safety and impressive benchmark performances—such as a record-setting score in programming and math challenges—yet it still demonstrates shortcomings on simpler tasks and requires expensive computational resources.

🔎 Google Preps Chatty AI Mode for Search
Google is developing an "AI Mode" for its search engine, offering chatbot-style interactions alongside traditional search results. This feature will display conversational answers with linked resources, enhancing the user experience. Positioned alongside tabs like "All" and "Images," it highlights Google's focus on integrating AI into search. With competitors blending chat and search, Google aims to maintain its lead in the evolving AI-powered search space.

💼 EU Clears NVIDIA's $700M Run:ai Acquisition
The European Commission has unconditionally approved NVIDIA's $700 million acquisition of Israeli AI startup Run:ai, determining it poses no competition concerns. Run:ai specializes in Kubernetes-based workload management and orchestration software for GPU-accelerated systems, complementing NVIDIA's hardware expertise. This strategic move aims to enhance NVIDIA's AI computing capabilities amid increasing regulatory scrutiny over its market dominance.

💰 Perplexity Secures $500M, Valued at $9B
Perplexity has raised $500M, reaching a $9B valuation, highlighting investor confidence in conversational AI search. The funding supports its innovation, including acquiring Carbon to enhance AI's connection to external data. Amid fierce competition with OpenAI and Google, Perplexity’s growth faces legal challenges over copyright, which could set industry standards. This positions the company as a pivotal player in the evolving AI-powered search landscape.

🎓 Arizona Approves AI-Only School, Sparks Heated Debate
Arizona has approved Unbound Academy, the first fully AI-taught online charter school, stirring debate on education's future. Supporters praise AI's potential to tailor learning and ease teacher shortages, while critics question its lack of human interaction and equity concerns. With curriculum details still vague, this bold initiative highlights both the promise and ethical risks of AI-driven education, potentially reshaping global learning models.

🛰 Ukraine Uses War Data to Train Combat AI
Ukraine is using over 2 million hours of drone footage to train AI for advanced battlefield applications. Curated by OCHI, the data enhances combat tactics, target identification, and weapon optimization. While allies express interest, the dataset's terrain specificity limits broader applicability. As both Ukraine and Russia increasingly adopt AI in warfare, this initiative highlights the technology's pivotal role in shaping modern military strategies.

🖋️ Palantir and Anduril Form Consortium to Bid on Defense Contracts
Palantir Technologies and Anduril Industries are leading a consortium, including companies like SpaceX and OpenAI, to compete for U.S. defense contracts, aiming to challenge traditional contractors such as Lockheed Martin and Boeing. This initiative seeks to leverage Silicon Valley's advanced technologies to provide more efficient and modern defense solutions.

☝️ POWERED BY LANGTRACE

Go from shiny demos to reliable AI products that delight your customers with Langtrace. Check out and star our GitHub for the latest updates and join the community of innovators.

20% discount for Langtrace here: https://langtrace.ai/matthewberman

👾 FORWARD FUTURE ORIGINAL

Econ 09 | AI, Energy, and Economic Efficiency

This is a series about the economics of AI. The economic question is somewhat related to the ecological question - the environmental impact. Let’s dig deeper into these related questions - energy, efficiency, economy and ecology / environment, for these are, I’m sure you’ll agree, very important. 

In conventional media, you’ve no doubt come across articles as to how energy-intensive a data center that runs AI is. While the underlying concern is very much valid, I’d propose that such a view in itself does not take into account the overall context; for a crucial point is, what is the corresponding resource consumption of the pre- or non-AI alternatives for the same output. 

So firstly, what does AI cost? Specifically, in our main case, Generative AI in the form of large language models (LLMs) that produce human-like output, which has been the focus of our discussion.  There are two main costs (and processes) in relation to this. Firstly, building these LLMs, and then running them for user access. We’ve discussed some of the details of this in my first article series on AI Concepts, but here’s a quick recap. 

Building an LLM entails, as a one-off, feeding a neural net (an algorithm capable of ‘learning’ patterns) large amounts of data (text), such as in the latest cases, practically all the content of the Internet, and subjecting it to further training and reinforcement, including that carried out by humans. To get to the scale of the latest LLMs and the emergent capabilities achieved at such scale, which is what makes Gen AI viable, this is a very energy-intensive process, the costs going to a few millions of dollars.  → Continue reading here.

✌️ POWERED BY EMERGENCE AI

Emergence AI, an agentic AI company, has launched its multi-agent orchestrator, designed to autonomously plan, execute, verify, and iterate in real time. This technology delivers advanced web automation for enterprises, allowing multiple AI agents to handle complex tasks like navigating dynamic menus, filling forms, and handling errors more effectively, such as page load failures, broken links, or unexpected pop-ups.

Emergence AI is inviting developers and enterprises to build and integrate their own agents under orchestration to solve their unique challenges. Reach out today via their Orchestrator API or email [email protected] for more information.

📽️ VIDEO

Microsoft CEO’s Shocking Prediction: “Agents Will Replace ALL Software"

In our latest video, Matt discusses Microsoft CEO Satya Nadella's vision of AI agents revolutionizing software by replacing traditional apps. With AI handling data and business logic, tools like Excel and SaaS layers may become obsolete. This marks a transformative shift in enterprise and consumer applications. Get the full scoop! 👇

🧰 TOOLBOX

AI Tools Crafting Stories, Designing Visuals, and Elevating Content Creation

Moodpaper | Emotions Wallpapers: Moodpaper uses AI to craft custom wallpapers based on your mood for a personalized design experience.

Zebracat AI | Social Media Videos: Zebracat AI turns text and audio into captivating videos, featuring avatars, voice cloning, and automated editing.

Sudowrite | Storytelling Mastery: Sudowrite helps writers craft stories with AI-powered brainstorming, editing, and plot-planning tools for enhanced creativity.

🗒️ FEEDBACK

Help Us Get Better

Reply to this email if you have specific feedback to share. We’d love to hear from you.

🤠 THE DAILY BYTE

Chat with AI Santa: Spreading Cheer and Funding Type 1 Diabetes Research

CONNECT

Stay in the Know

Follow us on X for quick daily updates and bite-sized content.
Subscribe to our YouTube channel for in-depth technical analysis.

Prefer using an RSS feed? Add Forward Future to your feed here.

Thanks for reading today’s newsletter. See you next time!

The Forward Future Team
🧑‍🚀 🧑‍🚀 🧑‍🚀 🧑‍🚀