In partnership with
Welcome back, AI enthusiast.
Microsoft just released Fara1.5, a browser agent that outperformed OpenAI and Google on a 300-task real-world benchmark.
It works across three model sizes and pauses to ask when something looks risky. Could this be the most underrated AI agent release of the year?
Today in the “Incredible AI” world:
Microsoft’s Fara1.5 Agents Crush Rivals
Trump’s AI Order Suddenly Pulled Back
Alibaba Built Marathon Coding Agents
How to Streamline Team Training with AI
5 Super-Useful AI Tools You Should Try
Top AI-Generated Image of the Day
Read Time: 4 minutes
- LATEST DEVELOPMENTS -
Microsoft Just Humiliated OpenAI and Google:
Microsoft just dropped Fara1.5, a family of browser AI agents that took OpenAI's Operator and Google's Gemini 2.5 out back and handed them a quiet loss. Three model sizes, one clear message. Microsoft is not playing around with browser AI anymore.
Things to Know:
What It Is: Microsoft Research released three agent sizes: 4B, 9B, and 27B. They watch browser screenshots and handle mouse and keyboard tasks on their own.
The Numbers: Fara1.5-27B hit 72% on the Online-Mind2Web benchmark. OpenAI's Operator scored 58.3%, and Gemini 2.5 Computer Use landed at 57.3%.
How It Thinks: The agent runs an observe-think-act loop at each step. It looks at the last three screenshots and either acts or stops to ask the user a question.
The Safety Angle: Fara1.5 pauses before any irreversible action. It also stops when details are missing or the task description is too unclear to act on.
Microsoft's Fara1.5 is fully open and outperforms both major paid alternatives right now. The predecessor, Fara-7B, scored only 34.1% on the same test. That jump tells you Microsoft put serious work into this one.
Advisers Derail Trump’s AI Order Plan:
Trump's big AI executive order had a signing ceremony ready, tech CEOs lined up, and cameras waiting. Then it all fell through hours before the event. Turns out, the president who doesn't want to regulate AI got cold feet about regulating AI.
Things to Know:
What Happened: The signing was called off after AI adviser David Sacks and several tech executives pushed back hard. Trump reportedly hated the regulation angle from the start.
Who Was Involved: Elon Musk, Mark Zuckerberg, and David Sacks all spoke with Trump the night before and the morning of the scheduled signing.
Why It Died: Trump told reporters he didn't like parts of the order. He said it would get in the way of America's lead over China in the AI race.
What's Still Unclear: The order would have set up a voluntary AI model testing program. Nobody knows when or if a revised version will come back to the table.
Advocates who had been pushing for AI safety measures in Washington thought this moment was finally happening. Now they're back to square one. If an AI order can't even survive its own signing ceremony, Washington might have a bigger problem than just AI.
This billionaire strategy beat the S&P 500 by 3.1x from 2017-2025. Here's how to get in.
That’s not by picking the right stock or timing the market. It’s by holding three real asset classes in one strategy.
It’s anchored by an investment typically exclusive to billionaires.
Could be good timing too–
Bloomberg's Marcus Ashworth recently wrote, there’s "no more reliable safe havens."
The S&P, while hovering over 5 year highs, fell over 7% from the February peak. Bonds might carry less risk but they are barely keeping pace with inflation.
The things supposed to protect your portfolio started moving together.
Meanwhile, the world's wealthiest have been setting records in postwar and contemporary art.
After the dot-com bust, it grew roughly 24% annually for a decade. After 2008, roughly 11% annually for 12 years.
It trades globally in multiple currencies, has scarce supply, and has shown near-zero correlation to equities since 1995.*
Masterworks has helped 70,000+ investors allocate $1.3B fractionally across 500+ artworks featuring Banksy, Basquiat, and Picasso.
See if you can improve your portfolio performance in one diversified strategy.
*According to Masterworks data. Investing involves risk. Past performance is not indicative of future returns. Important Reg A disclosures: masterworks.com/cd
Alibaba's New AI Worked for 35 Hours Straight:
Alibaba just released Qwen3.7-Max, and it did something worth paying attention to. The model ran completely on its own for 35 straight hours on a task no human could sit through. No hand-holding, no interruptions, no coffee breaks.
Things to Know:
The 35-Hour Test: Alibaba tasked the model with optimizing a complex attention kernel. It made 1,158 tool calls and ran 432 kernel evaluations without a single human stepping in.
The Results: By the end of the run, Qwen3.7-Max delivered a 10x speed improvement over the reference implementation. It redesigned the architecture five separate times to get there.
Cross-Harness Support: The model works natively with external agent frameworks like Claude Code and OpenClaw. Developers can plug it in without rebuilding their existing workflows from scratch.
The Catch: Unlike previous Qwen releases, this one is strictly API-only with no open weights. Pricing starts at $2.50 per million input tokens and $7.50 per million output tokens through Alibaba Cloud.
The open-source community that built its trust around Qwen is not thrilled about the closed-weights decision. That said, a model that independently rewrites hardware kernel code for 35 hours and scores top among Chinese AI models on independent benchmarks is hard to ignore, whatever the licensing terms say.
- TRY THIS -
How to Streamline Team Training with AI:
Keeping everyone on the same page at work can be tough. Processes get scattered, policies live in random docs, and new hires often learn by trial and error. What if you could centralize all of that into one simple system?
Trainual is an AI‑powered training and knowledge management platform that helps you document processes, assign training, and track progress—all in one place.
How it works:
Document your processes: Write down your company’s standard operating procedures (SOPs), policies, and workflows directly inside Trainual. No more scattered files—everything lives in one searchable hub.
Organize by roles and responsibilities: Map out who does what. Assign training paths based on job role, team, or location so each person sees only what’s relevant to them.
Build training modules: Turn your documentation into step‑by‑step lessons. Add text, images, or videos to make instructions clear and easy to follow.
Assign and track completion: Send training to individuals or teams. Progress bars and completion rates make accountability simple.
Test knowledge: Add quizzes or checkpoints to confirm that employees understand the material before moving on.
Keep everything updated: Collaborate with teammates to edit, comment, and approve changes. Updates are instantly reflected across the platform.
Integrate with your tools: Connect Trainual with your HR, payroll, or messaging apps so training fits seamlessly into your existing workflow.
Instead of piecing together scattered docs, Trainual gives your team one reliable source of truth. That means faster onboarding, fewer mistakes, and consistent training—every time.
Stop making AI decisions in the dark. Understand AI usage.
Leadership is asking: are we getting value from AI? Where are we exposed? Right now, most teams have no idea.
Harmonic Security automatically maps every AI interaction into the use cases driving real work — so CIOs can rationalize spend, CISOs get risk in context, and AI committees get proof of impact.
- BEST AI TOOLS -
MakeLogo AI: Generate custom logos and branding assets in minutes.
Robin AI: Review and draft legal contracts using AI-powered document analysis.
Yoodli: Practice public speaking with AI feedback on pacing, filler words, and clarity.
Maket: Generate architectural floor plans and layouts using AI design tools.
Promptable Sheets: Run AI-powered bulk operations and transformations inside spreadsheets.
- DAILY POLL AND RESULT -
Today’s Poll:
Q) Is AI improving decision-making speed?
Vote and find out the result tomorrow.
Yesterday’s Result:
Q) Should AI draft legal contracts for small businesses?
A) Yes, cost-effective help - 74% 👑
B) No, risk too high - 26%
The Best New App for Reading Newsletters:
Reading newsletters in the inbox is frustrating - it is noisy and easy to lose control of subscriptions.
That’s where Meco comes in as a savior.
Meco is a distraction-free app for reading newsletters outside the inbox. The app is packed with features designed to supercharge your learning from your favorite writers.
With Meco, you can enjoy your newsletters in an app built for reading, while giving your inbox space to breathe.
Add your newsletters in seconds and liberate your inbox today!
- AI IMAGE OF THE DAY -
A Flaming Blade Cuts Through the Darkness:
Prompt:
"A hyper-stylistic, highly detailed painting crafted from smeared red and black metallic paint. A charging samurai holding a burning katana races through a dark forest in streaked motion. Bold red strokes create energy and movement, while black accents add depth and dimension. Light and dark contrasts build tension and surreal dynamism. Masterpiece quality, bokeh lighting, intricate details."Try this prompt in any decent AI image generator and let me know your result.
- DO YOURSELF A FAVOUR -
If this email landed in your Promotions or Spam tab, move it to your Primary inbox. Takes 2 seconds, and you will never miss an edition again.
I put this together every day so you do not have to spend your morning hunting down AI news yourself. Would be a shame if it ended up buried under discount codes and bank alerts.
