In today’s world, being good at just one thing isn’t enough. You need to know how to think, build, sell, and adapt — all at once. That’s what I share in my AI newsletter and community. You’ll learn how to automate tasks, create products, grow businesses, and lead teams — using AI. I’ve worked hands-on across 5+ industries and almost every business role. Everything I share is simple, useful, and straight from experience. Join 5,000+ people learning how to stay sharp, flexible, and future-ready.
Share
✨Google rolls out Gemini 3 Flash
Published 25 days ago • 5 min read
The Creators Hive: AI News, Tools & Tips — Curated for Modern Creators and Innovators
Greetings Qrew,
Whether you're diving into AI for work, side projects, or just out of curiosity — this issue of The Creators Hive brings you practical insights, hands-on learning, and tools you can actually use.
Google has launched Gemini 3 Flash, a faster version of its flagship AI that still offers high-end reasoning and multimodal abilities. It is now the default model in the Gemini app and in Google Search’s AI Mode globally.
Key details:
Optimized for speed and low latency while keeping strong reasoning performance.
Set as the default model in the Gemini app, replacing earlier Flash versions.
Powers AI Mode in Search to answer complex, multi-part queries with structured responses.
Designed for everyday tasks like summarization, coding help, and image or video understanding.
Why it matters: This move shows Google is standardizing on a single fast frontier model for consumer-facing AI, not just a research system. It raises expectations for instant, more capable AI in search and apps, intensifying competition with OpenAI and others.
OpenAI is in early talks to raise up to 100 billion dollars in new funding at a potential valuation of around 830 billion dollars. The round would likely involve sovereign wealth funds and follow a previous valuation of about 500 billion dollars from a share sale.
Key details:
Target raise up to 100 billion dollars, reportedly by early 2026.
Implied valuation as high as 830 billion dollars if the full amount is raised.
Potential investors include large sovereign wealth funds due to the scale.
Company remains unprofitable and faces heavy compute and infrastructure costs.
Why it matters: If completed, this would make OpenAI one of the world’s most valuable private companies, approaching mega-cap tech territory. It signals that investors still see massive upside in foundational AI platforms despite high risks and costs.
Elon Musk told xAI staff that he believes the company could reach artificial general intelligence as early as 2026. At the same time, xAI is planning to scale its infrastructure from about 200,000 GPUs to over 1 million.
Key details:
Musk set a public internal goal of AGI around 2026, calling the next 2–3 years critical.
xAI’s “Colossus” project is expanding from roughly 200,000 GPUs to a target of 1 million.
Musk suggested access to 20–30 billion dollars per year in funding is possible for xAI’s efforts.
xAI is competing with OpenAI, Google, and Meta in large-scale model training and deployment.
Why it matters: These ambitions highlight how central massive compute buildouts are to the current AI race. They also raise questions about feasibility, power-grid impact, and whether AGI timelines are realistic or mainly motivational targets.
Figure AI CEO Brett Adcock is starting a new AI lab called Hark, reportedly funded with 100 million dollars of his own capital. The lab aims to build “human-centric” AI systems that are proactive, self-improving, and designed to care about people.
Key details:
Hark is a separate AI lab led by Adcock while he continues as Figure AI’s CEO.
Initial funding is about 100 million dollars in personal capital.
Focus is on human-centric AI that reasons proactively and improves over time.
The first GPU compute cluster for Hark has reportedly just gone online.
Why it matters: The move shows that even founders of major AI robotics firms see room for dedicated labs focused just on advanced AI systems. It underscores how capital and talent are fragmenting across multiple overlapping AI ventures, intensifying competition for GPUs and researchers.
The US FDA has qualified AIM-NASH, the first AI-based drug development tool to help assess liver disease in clinical trials. The AI system analyzes liver biopsy images to score fat, inflammation, and scarring, which can support faster testing of new drugs for metabolic liver disease.
Key details:
AIM-NASH is an AI tool for metabolic dysfunction-associated steatohepatitis (MASH) trials.
It quantifies steatosis, inflammation, and fibrosis in liver biopsy slides.
The tool is qualified by the FDA as a drug development tool, not as a standalone diagnostic.
Pathologists remain responsible for final interpretation, reviewing AI scores and images.
Why it matters: This is a concrete example of AI improving how new liver drugs are developed, beyond text or image generation. Regulatory acceptance of such tools could shorten trial timelines, reduce costs, and speed up treatments for serious liver disease.
Vocalremover.org is a free, browser-based audio tool that uses AI to separate vocals from the music in a song so you get both a karaoke-style instrumental and an isolated a cappella track. It is designed to be simple enough for beginners while still useful for musicians, producers, and creators.
What makes it unique?
Fully online, no installation or account needed for basic use.
Fast processing, often delivering separated tracks in a few seconds.
Generates two outputs at once: instrumental and isolated vocals.
Uses AI-based stem separation that preserves reasonable audio quality for most songs.
Use case magic
Create instant karaoke or backing tracks from favorite songs.
Extract clean vocals for remixes, mashups, and sampling.
Help singers and musicians practice by muting or soloing parts.
Adjust pitch and speed with its extra tools for rehearsals or creative edits.
Why it stands out
Vocalremover.org stands out because it delivers AI-powered vocal isolation in a single click, without accounts, paywalls, or complex audio software. For many casual creators and small studios, it offers a quick, low-friction way to get useful stems and practice tracks directly in the browser.
Masterpieces from the Hive 🎨
Here’s what our community has been creating this week using the Hive Wall:
Prompt: An anthropomorphic moose skiing; artsy, Disneyesque, hand-drawn.
Prompt: a realistic and fantastical painting, Santa Claus Express is portrayed with a dynamic, direct, and energetic identity. The scene conveys subtle motion through a light blur that suggests speed and action, enhanced by a warm color palette of ivory (#F6F1EB), soft beige, and gentle red accents. Santa stands confidently with his coat slightly floating, arms raised as if launching gifts, while golden light flares glow around his hands to express energy and movement. The background remains softly blurred, showing a cozy interior atmosphere with faint silhouettes of Christmas trees in the distance, reinforcing warmth and festive magic.
Prompt: Vast abandoned ancient city hidden under jungle overgrowth, massive stone structures and roads revealed by light beams, mysterious atmosphere, cinematic lighting, clear focal point on central ruin
Prompt: a urban explorer cellphone photography long exposition shot. a strong camera flash illuminates the pitch black dark. a italian 1500s renaissance style grey metal door shield cartouche door knocker with a backplate. the upper and lower halves of the backplate have long but thin turned urn finials. historical heritage european piece. ornate and bulky mouldings. dark blue grey color lacquered wood
Thanks for reading!
Got feedback? Reply to this email and let us know. We read everything and are constantly improving based on your suggestions.
🚀 Level up your creativity with Qolaba AI. Stop switching between tabs. We are the all-in-one startup platform that combines top-tier Chat, Image, Video, and Audio models in one place.
I build with AI and (Meditation), and help others do the same.
In today’s world, being good at just one thing isn’t enough. You need to know how to think, build, sell, and adapt — all at once. That’s what I share in my AI newsletter and community. You’ll learn how to automate tasks, create products, grow businesses, and lead teams — using AI. I’ve worked hands-on across 5+ industries and almost every business role. Everything I share is simple, useful, and straight from experience. Join 5,000+ people learning how to stay sharp, flexible, and future-ready.
The Creators Hive: AI News, Tools & Tips — Curated for Modern Creators and Innovators Greetings Qrew, Whether you're diving into AI for work, side projects, or just out of curiosity — this issue of The Creators Hive brings you practical insights, hands-on learning, and tools you can actually use. Musk says robot surgeons coming soon Apple chooses Google Gemini for Siri OpenAI acquires health startup Torch Deepgram hits $1.3 billion valuation Meta launches “Meta Compute” for massive AI...
The Creators Hive: AI News, Tools & Tips — Curated for Modern Creators and Innovators Greetings Qrew, Whether you're diving into AI for work, side projects, or just out of curiosity — this issue of The Creators Hive brings you practical insights, hands-on learning, and tools you can actually use. Gmail’s AI Inbox and new tools OpenAI’s dedicated ChatGPT Health Agentic AI in software delivery Midjourney’s new Niji 7 model Governments push back on Grok’s images Tool Recommendation of the Day:...
The Creators Hive: AI News, Tools & Tips — Curated for Modern Creators and Innovators Greetings Qrew, Whether you're diving into AI for work, side projects, or just out of curiosity — this issue of The Creators Hive brings you practical insights, hands-on learning, and tools you can actually use. Amazon brings Alexa+ to the web Nvidia’s open-source AI for self-driving cars Meta’s Chief AI Scientist Goes Full Scorched Earth Grok gets criminal investigation speedrun AI copyright wars enter boss...
