In partnership with
October 28, 2026 · READ ONLINE
Keeping you ahead in AI.
THE DAILY BRIEFING / 4 MIN READ
Welcome back, AI enthusiasts.
Google just unveiled Genie 4, an AI that can create entire virtual worlds you can explore and change while you're inside them. It raises a bigger question about how games, films, and even robots could be developed if creating realistic environments becomes as simple as describing them. Let’s dive in!
IN TODAY'S ISSUE
01 Google Unveils Genie 4 With AI Worlds That Feel Alive
02 Anthropic Is Now Developing Its Own Drugs With Claude
03 OpenAI drops New GPT 6.1 Terra
01 / AI VIRTUAL WORLD
Evolving AI: Google has unveiled Genie 4, an AI model that creates realistic virtual worlds users can explore and change in real time.
KEY POINTS
Genie 4 supports a few minutes of continuous interaction, compared with the earlier Project Genie prototype's 60-second limit.
Users can change weather, camera angles, and characters mid-scene, with more realistic human movement, collisions, and aerodynamics.
Google's first real-time world model with native audio generates environmental sounds, music, and interactive conversations.
DETAILS
Google DeepMind's Genie 4 lets people explore AI-generated worlds, change scenes as they move through them, and talk to the characters they meet. Gemini Omni helps create realistic visuals, while Gemini 3.8 Flash tracks the simulation and plans what happens next to keep events consistent. Early testers are exploring film scene previews and game prototypes, while Google demonstrated a simulated Apollo humanoid playing checkers. Google plans to bring Genie 4 to AI Ultra subscribers through Project Genie, giving more people the chance to create and explore their own interactive worlds, although no release date has been announced and longer simulations can still lose consistency.
WHY IT MATTERS
Google's Genie 4 could let people test game and film ideas simply by describing the worlds they want, reducing the time, cost, and technical skills needed to bring creative concepts to life. If this becomes a reliable workflow, Google could gain control over the earliest stages of creative production, putting pressure on traditional game engines and 3D software companies as creators increasingly turn to AI for prototyping. The bigger opportunity extends into robotics, where realistic virtual worlds could eventually allow AI agents and robots to practice complex tasks before attempting them in the real world, although simulation accuracy remains a major challenge.
Analytics on Live Data Without Leaving Postgres
When analytics on Postgres slows down, most teams add a second database. Then come the pipelines, the sync jobs, and a copy of your data that's always a little behind.
TimescaleDB takes a different approach: extend Postgres instead of splitting away from it. Hypertables partition your data automatically as volume grows. Hypercore compression cuts storage up to 95%. Continuous aggregates keep dashboards live without re-querying everything.
CERN runs Postgres this way for sensor data from the Large Hadron Collider.
No split architecture, no pipeline lag, no new query language to learn. Same SQL, same drivers, same tools.
Start on Tiger Cloud and get $1000 in credits.
02 / AI DRUG DEV
Source: The Guardian
Evolving AI: Anthropic is using Claude to develop experimental medicines, including new drug treatments of its own.
KEY POINTS
Claude generated about 35,000 antibody designs in a month, with 20 dislodging allergy-related antibodies in lab tests.
Claude helped shorten a cancer therapy component by 70%, producing tenfold more cancer-recognizing receptors in lab tests.
Anthropic bought a one-person biotech startup that used Claude to coordinate 11 labs and test a muscular dystrophy treatment in mice.
DETAILS
Anthropic is getting into drug development, working with six research partners and buying a one-person biotech startup to pursue treatments for rare diseases. By connecting Claude to scientific software and laboratory data, researchers can use it to design experimental molecules, improve drug candidates, interpret test results, review manufacturing records, and assess clinical trial plans while scientists handle physical testing. Although early results in allergy and cancer research are promising, most projects remain preclinical, with their safety and effectiveness in humans still unproven.
WHY IT MATTERS
Anthropic is becoming both an AI supplier to drug developers and a potential competitor through its own treatment programs, putting pressure on specialist AI drug-discovery companies. As Claude takes on more molecular design, experiment planning, and research analysis, these companies will need to prove that their proprietary data, specialized models, and laboratory capabilities deliver better results than a general-purpose AI. This competition could make early drug research cheaper and give smaller biotech teams more opportunities to develop treatments for rare diseases, although expensive clinical trials will still determine whether those treatments reach patients.
03 / MODEL LAUNCH
Evolving AI: OpenAI has launched GPT-6.1 Terra, a faster and more affordable AI model built to handle everyday computer tasks.
KEY POINTS
OpenAI reports Terra matches GPT-6 Astra on a professional computer-use benchmark while completing tasks over 3× faster.
Standard API prices are 50% lower than GPT-5.6 Terra and GPT-6.1 Sol, costing $1 per million input tokens and $5 per million output tokens.
Terra is rolling out to Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex, with access also available through the API.
DETAILS
OpenAI's GPT-6.1 Terra can look at what's on a computer screen and use software to get things done, whether that's filling out forms, processing invoices, editing videos, or working with technical drawings. In OpenAI's tests, the model matched GPT-6 Astra on professional computer tasks at medium reasoning effort, finished more than three times faster, and showed stronger coding results and fewer factual errors than its predecessor. It is available through ChatGPT Work, Codex, and the API at lower prices, although OpenAI still recommends its larger models for the most demanding tasks.
WHY IT MATTERS
OpenAI's GPT-6.1 Terra could lower the cost of automating everyday business tasks, making it more affordable for companies to process invoices, manage forms, and handle routine software work at scale. If businesses can use Terra to complete these jobs inside their existing applications, paying separate AI automation providers becomes harder to justify, especially when those providers rely on OpenAI's own models. This puts pressure on AI automation startups to lower prices and prove that their industry expertise, deeper integrations, and greater reliability can save businesses more money than using Terra directly, even after accounting for human review and corrections.
04 / HUMAN OR MACHINE?
One is real. Trust your eyes?
Two images. One photograph. Click the one you think is real.
Image 1
Image 2
05 / QUICK HITS
💰 OpenAI reportedly plans to buy Google's AI chips, potentially reducing its dependence on NVIDIA.
🤖 Google released Gemini Robotics Embodied Reasoning 2 in public preview, helping robots plan and execute complex tasks.
🇺🇸 Anthropic announced a 40% Claude discount for U.S. government agencies, estimating $1 billion in taxpayer savings.
🍎 OpenAI added iCloud Mail and Calendar connections to ChatGPT, enabling users to manage emails and schedules.
06 / TRENDING AI TOOLS
🤖 Lindy – The simplest way for businesses to create, manage, and share agents*.
📣 Needle - Finds leads and runs personalized outreach through your AI agent.
🎨 UseIt Studio - Turns ideas into editable, print-ready designs for physical products.
🎬 CutBay - Turns documents, links, slides or prompts into finished videos.
*PARTNER LINKS
BEFORE YOU GO
Worth your four minutes?
Help us make the next edition better.
See you tomorrow,
The Evolving AI team
