Updates and Insights for Today

1. MIT’s Radical New AI: Edit Images Just by Tinkering with Tokens!

2. Stargate Supercharge: OpenAI + Oracle Power Up 5 GW AI Beast!

3. Anthropic Signs EU Code: Safety First, Innovation Always!

4. The latest in AI tech

5. AI Tutorials: Google’s AI Mode Just Killed SEO—Here’s How to Survive.

6. AI tools to checkout

AI News

MIT’s Radical New AI: Edit Images Just by Tinkering with Tokens!

Quick Summary:
MIT scientists found a way to manipulate or generate images by adjusting compressed "tokens"—no traditional generator needed. This unlocks editing tools that can swap subjects, change poses, and even inpaint missing parts.

Key Insights:

• Researchers use a 1‑D tokenizer to compress images into a compact token sequence (like a hidden AI vocabulary).

• Individual tokens control visible traits: blur, resolution, pose, brightness etc.

• The team automated token editing and showed it's possible to turn a red panda into a tiger or a robin into a dog by optimizing tokens alone.

• They generate images with just a tokenizer + detokenizer + CLIP guidance—no standard image generator required.

Why It’s Relevant:
This marks a major shift in AI image tech. Editing and generation become lighter, cheaper, and faster—no massive model or GPU farms needed. That could empower creative tools on everyday devices. Also, it reveals a deeper understanding of how visual data is stored and changed in AI systems.

📌 Read More: MIT

Our Partner Today

The Only AI That Knows All Your Work

Most AI tools start from scratch every time. ClickUp Brain already knows the answers.

It has full context of all your work—docs, tasks, chats, files, and more. No uploading. No explaining. No repetitive prompting.

ClickUp Brain creates tasks for your projects, writes updates in your voice, and answers questions with your team's institutional knowledge built in.

It's not just another AI tool. It's the first AI that actually understands your workflow because it lives where your work happens.

Join 150,000+ teams and save 1 day per week.

Try ClickUp Brain Free.

Stargate Supercharge: OpenAI + Oracle Power Up 5 GW AI Beast!

Quick Summary:
OpenAI and Oracle have teamed up to boost Project Stargate with an extra 4.5 GW of data-center capacity—taking the total to over 5 GW and powering over 2 million AI chips. The expansion is expected to create 100,000+ jobs and strengthen U.S. leadership in generative AI infrastructure.

Key Insights:

• This new deal ups Stargate’s total U.S. capacity to more than 5 GW—surpassing the halfway mark toward the planned 10 GW over four years.

• OpenAI and Oracle expect to deploy 2 million+ chips across these data centers, with Oracle delivering Nvidia GB200 racks at the Texan Abilene site.

• The initiative will generate over 100,000 construction, operations, and indirect jobs—spanning roles from electricians to technicians.

• This also represents a $30 billion-per-year Oracle cloud commitment starting in 2028, and Oracle stock dipped ~2–3% on cost concerns.

Why It’s Relevant:
This expansion reaffirms U.S. dominance in AI infrastructure—offering immense computing power, job creation, and industrial revitalization. It also shows OpenAI diversifying beyond Microsoft by locking in Oracle as a major cloud partner. Though the wider Stargate project has faced earlier delays and scaled-back goals, this new deal signals serious momentum.

📌 Read More: OpenAI

Anthropic Signs EU Code: Safety First, Innovation Always!

Quick Summary

Anthropic, creators of Claude, announced on July 21, 2025 that they will sign the EU’s voluntary General‑Purpose AI Code of Practice. This aligns with their values of transparency, safety, and accountability—helping them stay compliant as the EU AI Act comes into effect on August 2, 2025. Reuters+14Anthropic+14LinkedIn+14

Key Insights

• Principled move—Anthropic joins OpenAI and likely Microsoft in backing EU standards for safety and transparency.

• Builds on existing policy—Extends their “Responsible Scaling Policy” and supports risk documentation, including chemical/biological threats.

• Economic motivation—EU projects a €1 trillion‑per‑year AI boost by the mid‑2030s; compliance ensures market access.

• Regulatory context—EU’s actual Code won’t be legally binding until late 2025, but signatories gain early “safe harbor” certainty.

Evaluation: Benefit or Constraint for EU Users?

Signing the Code is a net positive for EU users:

• ✅ Increased transparency on how Claude handles data, risks, and copyright

• ✅ Stronger safety measures, especially for powerful “frontier” models

• ⚠️ Potential trade-off: More rigorous documentation and audits might slow down new features Overall, sign-on means better user protections and clarity, with minimal impact on innovation. It’s more benefit than restriction.

Why It Matters
Anthropic’s move reinforces its global AI leadership and supports safer, trustworthy AI in Europe. By joining OpenAI and Microsoft, they're shaping compliance norms ahead of enforcement—while ensuring EU users enjoy clearer, safer AI tools without dampening innovation.

📌 Read More: Anthropic

 AI Tutorials

Google’s AI Mode Just Killed SEO—Here’s How to Survive

Quick Summary:

Google’s AI Mode is replacing traditional SEO with zero-click search answers. Websites are losing massive traffic as AI generates full summaries directly in results. But with the right strategy, businesses can still win—if they act fast.

Key Insights:

• AI Mode now shows up in 50%+ of Google searches—twice as often as six months ago.

• Organic traffic drops of 20% to 60% are hitting sites overnight due to AI summarization.

• Results in AI Mode are unstable—91% change across repeat searches.

• Websites with authority + deep content (e.g., Wikipedia, YouTube, Reddit) dominate AI citations.

What Can I Learn?

• Why AI Mode is replacing 10 blue links with summarized answers

• How unpredictable AI results disrupt classic SEO tactics

• Why being “comprehensive” beats keyword targeting

• What content formats AI prefers (video, structure, schema)

Which Benefits Do I Get?

• Learn how to future-proof your SEO strategy

• Discover how to get cited by AI even without top rankings

• Understand how to structure content for AI visibility

• Gain tools, community, and free courses to act immediately

Why It Matters

These three breakthroughs represent the full spectrum of AI’s global impact. MIT’s token-based image editing redefines how we create visuals. OpenAI and Oracle’s Stargate expansion brings industrial-scale compute power to life. Anthropic’s EU Code signature anchors AI safety in real-world regulation. Together, they show how innovation, infrastructure, and integrity are shaping the AI era—right now.

Here is the full Video Tutorial 👉 Click Here

The latest in AI tech

🐝 Amazon Acquires Bee for $50 AI Wristband

Amazon is snapping up Bee, a San Francisco startup whose $50 wristband listens to conversations, then generates reminders, summaries, and to‑dos. Privacy controls are being added to mute recording, with the team joining Amazon Devices. This marks Amazon’s latest move into AI wearables, following its Halo exit, and signals growing focus on personal voice assistants.
📌 Read More: CNBC

✈️ Delta’s AI Ticket Pricing Sparks Lawmaker Alarm

Delta plans to use AI-powered pricing (via Fetcherr) on 20% of domestic flights by end of 2025, up from 1%. Officials and senators are raising concerns about “predatory” dynamic fares and privacy. Delta insists it won’t use personal data and that AI enhances existing dynamic pricing. But US lawmakers are demanding transparency and fairness safeguards.
📌 Read More: Economic Times

🇬🇧 OpenAI & UK Seal Strategic AI Security Pact

OpenAI and the UK government signed a deal on July 21, 2025 to deepen AI security research and invest in UK data‑center capacity. The agreement includes sharing insights with the UK AI Security Institute and expanding OpenAI’s London branch. This move supports the UK’s billion‑pound AI strategy and ambition to become a global AI powerhouse.
📌 Read More: Reuters

🏗️ Anthropic’s ‘Build AI in America’ Energy Blueprint

Anthropic published a “Build AI in America” report urging the US to mobilize at least 50 GW of new power capacity by 2028 for frontier AI. Recommendations include faster permitting, grid upgrades, and workforce investment. The goal: keep AI development and jobs domestic, enhancing economic and national security amidst China’s rapid build-out.
📌 Read More: Anthropic

Our Second Partner Today

Marketing ideas for marketers who hate boring

The best marketing ideas come from marketers who live it. That’s what The Marketing Millennials delivers: real insights, fresh takes, and no fluff. Written by Daniel Murray, a marketer who knows what works, this newsletter cuts through the noise so you can stop guessing and start winning. Subscribe and level up your marketing game.

Sign Up Free

 AI Tools to check out

1. CZero: Your personal AI interface that works with any cloud AI provider while
   preserving your unique value in anAI-first world.

2. Rumora: Customers from  YouTube Without Running Ads.

3. FastBots: Automatically handle customer questions, capture leads, and support visitors across your site and social platforms.

4. Inliner: Create stunning visuals by simply describing what you want in a URL. (Free start)

Thanks for sticking with us to the end!

We'd love to hear your thoughts on today's email!

Your feedback helps us improve our content

⭐⭐⭐Superb
⭐⭐Not bad
⭐ Could've been better

Not subscribed yet? Sign up here and send it to a colleague or friend!

See you in our next edition!

Tom