Learn from this investor’s $100m mistake

In 2010, a Grammy-winning artist passed on investing $200K in an emerging real estate disruptor. That stake could be worth $100+ million today.

One year later, another real estate disruptor, Zillow, went public. This time, everyday investors had regrets, missing pre-IPO gains.

Now, a new real estate innovator, Pacaso – founded by a former Zillow exec – is disrupting a $1.3T market. And unlike the others, you can invest in Pacaso as a private company.

Pacaso’s co-ownership model has generated $1B+ in luxury home sales and service fees, earned $110M+ in gross profits to date, and received backing from the same VCs behind Uber, Venmo, and eBay. They even reserved the Nasdaq ticker PCSO.

Invest for $2.90/Share

_{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:

• AI World Divided — GPT-5 Debate Shows Deeper Fault Lines

• Qwen’s New Image Tool Swaps, Spins, and Styles with Accuracy

• Foxconn to Run SoftBank’s Stargate AI Server Plant in Ohio

• Meta & Character.AI Face Child Safety Probe

• Nvidia Launches Nemotron-Nano-9B-v2

The GPT-5 Debate Is Insane

A heated clash unfolded online between an AI researcher and a popular YouTuber over GPT-5’s performance and the broader state of AI. One pushed skepticism and credentials; the other championed optimism and public engagement. 

The argument spiraled into personal attacks, apologies, blocks, and meme wars. At its core, the drama reflects deeper divides in the AI world—about progress, trust, and who gets to shape the narrative of what intelligence really means.

WATCH THE VIDEO ON YOUTUBE

Qwen’s New Image Tool Swaps, Spins, and Styles with Accuracy

Qwen-Image-Edit is an upgraded model that lets anyone fix or transform pictures with ease. It can change objects, spin them around, swap styles, add or erase parts, and edit text in English or Chinese while keeping fonts intact. The system pairs a vision model and a shape-model to control both meaning and looks, so edits stay true to the scene. Tests show it leads all tools in accuracy and detail.

Why it matters

1. Brings pro-level editing to everyone – Simple prompts replace complex software, shrinking the skill gap.
   

2. Fuses image and text control – Seamless handling of words and visuals points to richer, truly multimodal models.
   

3. Raises performance bar – State-of-the-art results push other labs to improve, speeding progress across creative AI tools.

Foxconn to Run SoftBank’s Stargate AI Server Plant in Ohio

SoftBank will convert Foxconn’s former Ohio EV plant into a hub that builds AI servers for its $500 billion Stargate project with OpenAI and Oracle. Foxconn, under its Hon Hai name, keeps running the site, while SoftBank supplies new gear and funds. A joint venture will make equipment for data centers, marking the first concrete factory in Stargate’s US manufacturing push and signaling deeper US-Asia ties in the AI hardware race.

Why it matters

1. Secures vital hardware supply – US-made servers reduce reliance on overseas factories, easing chip and server shortages for AI firms.
   

2. Marries Asian know-how with US demand – Foxconn’s manufacturing skill plus SoftBank-OpenAI capital accelerates large-scale AI infrastructure.
   

3. Signals rising mega-projects – A $500 billion buildout shows investors’ confidence in AI growth and pressures rivals to match such massive bets.

Meta & Character.AI Face Child Safety Probe

Texas Attorney General Ken Paxton is investigating Meta’s AI Studio and Character.AI for allegedly pitching their chatbots as mental-health helpers without real medical backing. His probe says the bots lure kids with therapist-like personas, log conversations, and mine data for ads while offering generic advice. Both firms post disclaimers, yet children may ignore them. Paxton issued legal orders for documents to see if the companies broke state consumer-protection laws, rules.

Why it matters

1. Regulatory wake-up call — Governments are scrutinizing mental-health chatbots, forcing clearer claims and professional oversight.
   

2. Privacy spotlight — Investigators focus on how chat logs feed ad systems and training data, pushing stricter data-handling norms.
   

3. Child-safety design pressure — Developers must add robust age gates and crisis-escalation features, shaping the next wave of empathetic AI.

🧠RESEARCH

DINOv3

DINOv3 is a powerful self-learning vision model that works without labeled data. It scales well with big datasets, avoids quality loss during long training, and adapts easily to different tasks. It beats older models in many image tasks and offers flexible, high-performance tools for various real-world uses.

SSRL: Self-Search Reinforcement Learning

SSRL (Self-Search Reinforcement Learning) shows that large language models (LLMs) can act as their own search engines for reinforcement learning tasks. By training models to use internal knowledge instead of external tools, SSRL boosts accuracy, cuts costs, and reduces errors. It offers a stable, scalable path for smarter AI agent training.

Thyme: Think Beyond Images

Thyme is a new AI model that teaches multimodal language models to “think beyond images” by running code to manipulate visuals and solve problems. It crops, rotates, enhances images, and does math — all on its own. With two-stage training and a custom RL algorithm, Thyme boosts reasoning and perception accuracy across tough benchmarks.

🛠️TOP TOOLS

Rose AI - Cloud-based data platform that revolutionizes financial data analysis and visualization through the power of generative AI

DupDub - AI-powered content creation platform designed to streamline the production of multimedia content

Gliglish - AI-powered language learning platform designed to help users improve their speaking and listening skills through interactive conversations

ClassPoint AI - Instantly generate engaging quiz questions from slide content

MyVocal AI - AI-powered platform that offers rapid voice cloning and text-to-speech capabilities

AI Tool Spotlight

This AI Agent Replaces ChatGPT, Notion & Manus (Tested) | Skywork

Skywork is a powerful AI tool that acts like a team of digital assistants. It creates documents, slides, websites, spreadsheets, and even podcasts—all based on deep research or uploaded data. It automates tasks like analyzing spreadsheets, building reports, and generating presentations. The output is clean, accurate, editable, and fully cited. From business analytics to content creation, Skywork saves time while delivering high-quality, human-level results in minutes.

🗞️MORE NEWS

• Nvidia released Nemotron-Nano-9B-v2, a compact AI model that fits on a single A10 GPU, supports multiple languages, and lets users toggle reasoning on or off. It's fast, accurate, open-source, and ready for commercial use.

• Grammarly has launched nine free AI writing agents that help students write, edit, and even predict paper grades. Tools include a grader, proofreader, paraphraser, citation finder, and plagiarism checker, aiming to boost skills, not replace learning.

• Meta's upcoming smart glasses, Hypernova, will reportedly cost around $800—much lower than the $1,300 estimate. With a built-in display, photo features, and gesture controls, they aim to bring smart eyewear to more people.

• AI startup Perplexity now offers live transcripts of quarterly earnings calls for Indian public companies. The Finance dashboard already featured U.S. stock data, market news, watchlists, charts, and crypto tracking, this new update adds India-focused investor tools and call schedules.

• A Google Cloud survey shows 87% of game developers now use AI agents to speed up tasks, cut costs, and focus on creativity. However, many worry about job losses, ownership issues, and high integration costs.

• OpenAI CEO Sam Altman warns the U.S. is underestimating China’s rapid AI progress, especially in open-source models like DeepSeek. He doubts export controls are effective and says China’s advancements prompted OpenAI to release its own open-weight models.

• DoorDash is scaling AI across teams to automate tasks, personalize learning, and empower non-engineers to build tools. AI enhances HR workflows, speeds up feedback analysis, and helps create tailored development plans—without replacing human judgment.