Good morning, { AI enthusiasts }. It took lower than every week for OpenAI’s Pentagon deal to go from announced to rewritten, and Sam Altman is asking the fallout “really painful.”
With protests outside the corporate’s HQ, mass cancellations, and its detailed contract revisions bringing as many questions as answers, the rushed agreement is popping what was presupposed to be an influence move into OpenAI’s biggest brand crisis in years.
In today’s AI rundown:
-
OpenAI walks back Pentagon details after backlash
-
ChatGPT upgrade fixes the ‘cringe’ problem
-
Create killer thumbnails with Midjourney
-
Google’s recent 3.1 Flash-Lite pairs speed, cost, intelligence
-
4 recent AI tools, community workflows, and more
LATEST DEVELOPMENTS
OPENAI
Image source: Sam Altman on X
The Rundown: OpenAI CEO Sam Altman just posted a note on X detailing significant revisions to the corporate’s initial Pentagon contract, coming amid worker pushback, user cancellations, and a surge of sign-ups to Anthropic following the deal.
The small print:
-
OAI’s original agreement used the identical language Anthropic had refused, finalized inside 24 hours following the Pentagon’s ban of its rival.
-
Altman called the deal rushed and said it looked “opportunistic and sloppy,” adding that he would “reasonably go to jail” than follow an unconstitutional order.
-
Research scientist Noam Brown clarified that OAI “won’t be deploying to the NSA or other DoW intelligence agencies for now,” as loopholes are addressed.
-
Altman held an all-hands on Tuesday, calling the deal “complex but the precise decision with extremely difficult brand consequences and negative PR for us.”
Why it matters: It’s been quite the headache for OAI post-Pentagon deal, with Altman admitting the corporate moved too fast — and consequences starting from bad optics to massive consumer backlash and protests outside its SF offices. The amended contract language is a start, however the brand damage seems like it’s already been done.
TOGETHER WITH VISA
The Rundown: The trust layer for agentic commerce. Visa’s Trusted Agent Protocol allows AI agents to confirm who they’re, who they represent, and what they’re authorized to do — so merchants can interact with agents with confidence, and humans stay on top of things.
Visa’s Trusted Agent Protocol helps deliver:
-
Cryptographic verification of AI agent identity and intent
-
Clear authorization and accountability for agent‑initiated actions
-
Trust signals to assist merchants distinguish legitimate agents from bad actors
Learn more here.
OPENAI
Image source: OpenAI
The Rundown: OpenAI just released GPT-5.3 Easy, overhauling its default ChatGPT model to all users with a brand new release that prioritizes how the AI talks over the way it thinks — tackling the preachy personality that’s frustrated users for months.
The small print:
-
OAI built 5.3 Easy around conversational quality, dialing back unnecessary refusals and what the corporate itself called a ‘cringe’ tone in responses.
-
The update also claims a discount in hallucinations, with rates down by over 25% on web search and nearly 20% on internal knowledge on benchmarks.
-
OAI says 5.3 Easy can be a “stronger writing partner,” moreover providing higher web answers and presentation of knowledge.
-
The corporate also teased a more significant model jump coming soon, posting an easter egg on X saying “5.4 before you Think.”
Why it matters: Anthropic’s Super Bowl ads were controversial, but their poking at ChatGPT’s personality can have cut a bit too near the reality — with OAI’s recent rollout prioritizing moving its personality away from the “cringe” tone. But with uninstalls up 295% after the Pentagon drama, personality is probably not essentially the most pressing issue.
AI TRAINING
The Rundown: Learn a fast prompting and editing system to make thumbnails that stand out, but don’t seem like AI slop. Whether it’s to your YouTube channel, a blog, and even your social feeds, learning to quickly make good thumbnails pays dividends.
Step-by-step:
-
Open Claude or any chat agent. Tell it your video topic and ask it to generate 4 Midjourney prompts with this structure: [Person + Expression] +[Action/Prop] + [Setting] + [Lighting] + [Composition with negative space] –ar 16:9.
-
Paste each prompt into the Midjourney web app. While you see a generation you want, click on it and check out choosing vary > subtle or vary > strong until you get the proper variation. Save the image to your desktop.
-
Key step: the design. Remember, simplicity is important to this system. Add 1–5 words of massive, click-worthy text in Canva.
-
Bring the opacity down 5% and use a blocky font. Go along with white or yellow text and optionally a black outline.
Pro tip: In the event you can’t get the text to suit right in your image, tell Claude to give you a brand new prompt that may create that vacant space for you.
PRESENTED BY GLEAN
The Rundown: Join Glean’s virtual Security Showcase on March 12 to shut the gap between AI ambitions and security readiness. Get a practical blueprint for governing AI agents at scale from the leaders constructing, securing, and deploying them in production.
Register for Glean’s Security Showcase to:
-
Hear Cvent’s CIO and CISO in a hearth chat on real-world AI agent risk decisions.
-
Get access to a brand new security framework, designed for governed AI agents at scale.
-
See what’s recent in agent controls, data protection, and personal‑by‑design deployment for enterprise AI.
Image source: Google
The Rundown: Google just rolled out Gemini 3.1 Flash-Lite, the corporate’s fastest entry in its Gemini 3 lineup that gives a near-instantaneous feel and upgraded intelligence while undercutting rivals on price.
The small print:
-
Flash Lite rounds out Google’s tiered Gemini 3 release weeks after Pro, giving a budget option for high-volume work that doesn’t need a flagship model.
-
Lite scored a 12-point jump on the Artificial Evaluation Intelligence Index over its predecessor, beating even larger prior-gen Gemini models on reasoning.
-
The model costs 1/4 of Anthropic’s Haiku and 1/eighth of Gemini 3.1 Pro, though output pricing triples from the two.5 version it replaces.
Why it matters: Low cost, fast models have gotten the true battleground in AI, and Flash-Lite’s benchmarks suggest Google isn’t sacrificing much intelligence to get there. But for all of the benchmark strengths that the Gemini 3 has dropped at the table, its consumer impact hasn’t felt on the identical level as Anthropic and OpenAI in 2026.
QUICK HITS
-
👷 Viktor – OpenClaw for enterprise with 3,000+ integrations in Slack and Microsoft Teams, delivering board-ready PDFs, dashboards, and web apps*
-
⚡️ Gemini 3.1 Flash-Lite – Google’s fastest, most cost-effective Gemini 3 model for high-volume dev workloads
-
🧠 GPT-5.3 Easy – OAI’s ChatGPT default model update with fewer refusals and fewer hallucinations
-
📚 NotebookLM – Recent custom styles for infographics to show info into visuals
Alibaba’s Qwen team faced a wave of exits, as staffers posted a coordinated “Qwen is nothing without its people” message echoing OpenAI’s 2023 mutiny.
Cursor CEO Michael Truell said its AI agent autonomously solved an open math research problem over 4 days, with stronger results than the official human solution.
Anthropic reportedly submitted a proposal for a $100M Pentagon drone swarm challenge before being barred from DoD work, as rival firms were chosen as an alternative.
xAI released a brand new ‘Beta 2’ version of Grok 4.20, with the update featuring higher instruction following, reduced hallucinations, and more.
OpenAI VP of Research Max Schwarzer announced he’s joining Anthropic, saying he’s “looking forward (to) supporting my friends there at this vital time.”
COMMUNITY
Every newsletter, we showcase how a reader is using AI to work smarter, save time, or make life easier.
Today’s workflow comes from reader Lenore B. in Recent Zealand:
“I had to arrange for a hearing where I can be cross-examined. I uploaded the documents (with confidential details removed) into an AI project and instructed it to be demanding and skeptical. I also asked it to review each session and discover what I used to be doing well. It stopped unproductive rumination, sharpened my reasoning, and gave me structured rehearsal. I walked into the hearing calm, clear, and assured.”
How do you utilize AI? Tell us here.
-
Read our last AI newsletter: Supreme Court geese AI copyright query
-
Read our last Tech newsletter: Apple’s budget iPhone goes AI-first
-
Read our last Robotics newsletter: Bio-robotic spy roaches
-
Today’s AI tool guide: Create killer thumbnails with Midjourney
-
RSVP to our next workshop on March 12: AI coding bootcamp pt. 1
That is it for today!
- ⭐️⭐️⭐️⭐️⭐️ Nailed it
- ⭐️⭐️⭐️ Average
- ⭐️ Fail
Login or Subscribe to participate
See you soon,
