Keep This Stock Ticker on Your Watchlist

They’re a private company, but Pacaso just reserved the Nasdaq ticker “$PCSO.”

No surprise the same firms that backed Uber, eBay, and Venmo already invested in Pacaso. What is unique is Pacaso is giving the same opportunity to everyday investors. And 10,000+ people have already joined them.

Created a former Zillow exec who sold his first venture for $120M, Pacaso brings co-ownership to the $1.3T vacation home industry.

They’ve generated $1B+ worth of luxury home transactions across 2,000+ owners. That’s good for more than $110M in gross profit since inception, including 41% YoY growth last year alone.

And you can join them today for just $2.90/share. But don’t wait too long. Invest in Pacaso before the opportunity ends September 18.

Invest While You Still Can

_{Paid advertisement for Pacaso’s Regulation A offering. Read the offering circular at invest.pacaso.com. Reserving a ticker symbol is not a guarantee that the company will go public. Listing on the NASDAQ is subject to approvals.}

Today:

• OpenAI Signs $300B Cloud Deal With Oracle

• Mira Murati’s $12B Lab Aims to Make AI More Consistent

• Anthropic Outage Briefly Knocks Claude, Console, and APIs Offline

• Grammarly Now Speaks 5 New Languages

• Google Releases Genkit Go 1.0 — AI Framework for Go Developers

OpenAI Signs $300B Cloud Deal With Oracle

OpenAI agreed to buy $300 billion of computing from Oracle over five years, dwarfing its present income. The contract demands 4.5 gigawatts—about two Hoover Dams—of electricity for AI data centers. Oracle’s disclosed future revenue jump sent its share price up 43 percent; investors cheered the bold long-term bet.

KEY POINTS

• Record-setting cloud deal: One of the biggest technology contracts ever, showing huge demand for AI computing power.
  

• Massive energy need: 4.5 gigawatts is enough to light four million homes, stressing the power grid.
  

• Market impact: Oracle’s stock soared after revealing $317 billion in future contract revenue, mostly from this deal.

Why it matters

OpenAI needs far more computers to train smarter AI. Oracle gets steady income and a big boost to its reputation. The size of the deal signals that spending on AI hardware and electricity is rising fast, affecting energy use, tech competition, and investors.

Mira Murati’s $12B Lab Aims to Make AI More Consistent

Thinking Machines Lab, led by ex-OpenAI chief Mira Murati, revealed research on making AI answers repeatable. By controlling tiny programs inside graphics chips, the team hopes to stop random output and boost learning. Their first product, due soon, may use this method for businesses and researchers.

KEY POINTS

• Chip timing fix: Finds that split-second GPU timing makes answers vary and proposes controls to lock outputs.
  

• Smoother training: Steady replies clean up reward signals, helping tailor models for companies.
  

• Open glimpse: First public post from the $12 billion lab signals a vow of open research before its debut product.

Why it matters

Consistent answers make AI easier to trust and teach. If the lab solves this, firms gain steadier tools and scientists get clearer tests. It also shows newcomers, not just tech giants, can push AI progress.

Anthropic Outage Briefly Knocks Claude, Console, and APIs Offline

Anthropic’s Claude chatbot, developer Console, and APIs went offline Wednesday, triggering a brief panic among users. Reports surfaced around 12:20 p.m. ET; service returned minutes later. The company blamed a short glitch before 9:30 a.m. PT. Recent recurring bugs highlight reliability challenges for the fast-growing AI provider.

KEY POINTS

• Service disruption: Claude, Console, and APIs were down; fixes were rolled out within minutes.
  

• User reaction: Developers joked online about having to code without AI help during the outage.
  

• Pattern of issues: Adds to a series of recent bugs that raise questions about Anthropic’s stability.

Why it matters

Reliable tools are critical for businesses that build on Claude. Even short outages can halt work and erode trust. Repeated glitches remind the AI industry that rapid growth must be matched with solid infrastructure and clear communication when things go wrong.

🧠RESEARCH

Parallel-R1: Towards Parallel Thinking via Reinforcement Learning

Parallel-R1 is a new training method that teaches AI to think in parallel—trying multiple ideas at once. It starts with simple examples, then uses reinforcement learning for harder problems. This approach boosts performance on tough math tests, improving accuracy by over 40% on some benchmarks compared to older methods.

Visual Representation Alignment for Multimodal Large Language Models

VIRAL is a new training method that helps AI models better understand visual details by aligning their vision components with expert vision models. This boosts performance in image-focused tasks like object counting and spatial reasoning. The approach shows consistent gains across benchmarks and suggests a new path for improving multimodal AI.

Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search

Mini-o3 is a new AI system designed for tough visual search tasks. Unlike earlier models, it supports many rounds of reasoning and trial-and-error thinking. Using a custom dataset and smart training tricks, it mimics OpenAI’s o3 model and improves accuracy as interactions deepen—showing stronger reasoning the more it thinks.

🛠️TOP TOOLS

Each listing includes a hands-on tutorial so you can get started right away, whether you’re a beginner or a pro.

ImgCreator AI — AI Art Image Generator - Text to image, image to image & Background Changer

Fal AI - AI Image & Video Generator - Develop and fine-tune models with serverless GPUs and on-demand clusters.

OpenAI FM – AI Voice Generator - An interactive demo for developers to try the new text-to-speech model in the OpenAI API.

📲SOCIAL MEDIA

🗞️MORE NEWS

• Grammarly now helps users write better in Spanish, French, Portuguese, German, and Italian. It edits grammar, rewrites for clarity, and offers translations—all inside the app. AI powers this major multilingual upgrade.

• Google launched Genkit Go 1.0, a stable AI framework for Go developers. It supports major models, tool calling, and agent workflows. The new init:ai-tools boosts coding with integrated AI assistant features.

• Oracle launched its AI Center of Excellence for Healthcare to help hospitals use AI effectively. The center offers expert support, secure cloud tools, best practices, and training to boost care, research, and operations.

• YouTube's multi-language audio dubbing feature is now available to all creators. Powered by Google's Gemini AI, it boosts global reach and engagement. Localized thumbnails are also being tested to support international audiences.

• Harvard researchers developed PDGrapher, a free AI tool that finds treatments reversing disease in cells. It maps complex gene networks, predicts drug targets, and speeds drug discovery—outperforming other models in accuracy, speed, and insight.

• Microsoft AI chief Mustafa Suleyman warns that simulating consciousness in AI is dangerous. While emotional understanding is important, designing AI to appear sentient risks misleading users and could prompt misguided calls for AI rights.

• Oracle co-founder Larry Ellison gained $200 billion this year from a surge in AI-related stock, putting him close to Elon Musk’s net worth. Oracle shares soared while Tesla's drop cut into Musk’s lead.