WHAT’S AT STAKE TODAY ⚡

• OpenAI launches two 'open' AI reasoning models 🤖
• Trump says he'll announce semiconductor and chip tariffs 💻
• Google says its AI-based bug hunter found 20 security vulnerabilities 🔍
• AI-powered fintech Alaan raises $48M, one of the largest Series A rounds in MENA 💰
• US adds OpenAI, Google, and Anthropic to list of approved AI vendors for federal agencies 🏛️
• DeepMind thinks its new Genie 3 world model presents a stepping stone toward AGI 🧠
• ElevenLabs launches an AI music generator, which it claims is cleared for commercial use 🎵
• Three weeks after acquiring Windsurf, Cognition offers staff the exit door 🚪

OpenAI releases first open-weight AI reasoning models in five years.

OpenAI launches two ‘open’ AI reasoning models

OpenAI has launched its first open-weight AI models in over five years, releasing two reasoning models with capabilities similar to its proprietary o-series. The gpt-oss-120b and gpt-oss-20b models are now freely available on Hugging Face, marking a significant shift in OpenAI's strategy.

The larger gpt-oss-120b model can run on a single Nvidia GPU, while the lighter gpt-oss-20b operates on consumer laptops with 16GB of memory. Both models can send complex queries to cloud-based AI systems, allowing developers to connect them to OpenAI's more capable closed models when needed for tasks like image processing.

This release represents a dramatic departure from OpenAI's recent closed-source approach. CEO Sam Altman acknowledged in January that the company had been "on the wrong side of history" regarding open sourcing. The move comes amid intense pressure from Chinese AI labs like DeepSeek, Alibaba's Qwen, and Moonshot AI, which have dominated the open-source AI space with highly capable models.

The Trump administration's push for U.S. AI developers to open source more technology to promote American-aligned AI values also influenced this decision. "We are excited for the world to be building on an open AI stack created in the United States, based on democratic values," Altman stated.

Benchmark results show the models performing competitively against leading open-source alternatives. On Codeforces competitive coding tests, gpt-oss-120b scored 2622 and gpt-oss-20b scored 2516, outperforming DeepSeek's R1 while falling short of OpenAI's proprietary o3 and o4-mini models. On Humanity's Last Exam, they achieved 19% and 17.3% respectively, again surpassing other open models but underperforming OpenAI's closed systems.

However, the open models suffer from significantly higher hallucination rates. On OpenAI's PersonQA benchmark, gpt-oss-120b and gpt-oss-20b hallucinated in 49% and 53% of responses respectively, compared to just 16% for the o1 model and 36% for o4-mini. OpenAI attributes this to smaller models having less world knowledge than larger frontier systems.

Both models utilize mixture-of-experts (MoE) architecture for efficiency. The gpt-oss-120b, despite having 117 billion total parameters, only activates 5.1 billion parameters per token. They were trained using high-compute reinforcement learning, similar to the o-series, enabling a chain-of-thought reasoning process that takes additional time and computational resources.

This training approach makes the models particularly effective for powering AI agents, with capabilities including web search and Python code execution within their reasoning process. However, they are text-only models, lacking the multimodal capabilities of OpenAI's other offerings.

The models are released under the permissive Apache 2.0 license, allowing enterprises to monetize them without paying OpenAI. However, unlike fully open-source alternatives, OpenAI won't release the training data, likely due to ongoing copyright litigation concerns.

OpenAI delayed the release multiple times to address safety concerns. Beyond standard safety policies, the company investigated whether bad actors could fine-tune the models for cyberattacks or creating biological/chemical weapons. While testing revealed marginal increases in biological capabilities, neither model reached OpenAI's "high capability" danger threshold.

The release positions OpenAI to compete in the increasingly important open-source AI market while maintaining its commercial closed-model business. However, the company faces continued competition from upcoming releases, including DeepSeek's anticipated R2 model and new offerings from Meta's Superintelligence Lab.

This strategic pivot reflects the evolving AI landscape, where open-source models are becoming crucial for developer adoption and geopolitical influence, forcing even commercially-focused companies like OpenAI to embrace more open approaches.

Trump announces upcoming semiconductor tariffs amid ongoing chip manufacturing challenges.

Trump says he’ll announce semiconductor and chip tariffs

President Trump announced plans to impose semiconductor tariffs as early as next week, though details remain unclear. This could significantly disrupt U.S. hardware and AI companies, despite America producing only 10% of global chips. While the 2022 CHIPs Act allocated $52 billion for domestic manufacturing and companies like TSMC committed $100 billion investments, scaling production faces delays. Intel recently postponed its Ohio facility construction again. The announcement coincides with pending decisions on AI chip export restrictions, as the administration debates whether to replace Biden's country-specific export rules with new policies.

⚡ More AI Bites

1. 🔍🤖 Google says its AI-based bug hunter found 20 security vulnerabilities. Robot debugger scores big security wins!
2. 💰🏦 AI-powered fintech Alaan raises $48M, one of the largest Series A rounds in MENA. Middle East fintech gets massive funding boost!
3. 🏛️✅ US adds OpenAI, Google and Anthropic to list of approved AI vendors for federal agencies. Big AI names get government stamp of approval!
4. 🌟🧠 DeepMind thinks Genie 3 world model presents stepping stone towards AGI. Google's AI lab claims major breakthrough in general intelligence!
5. 🎵🤖 ElevenLabs launches an AI music generator, which it claims is cleared for commercial use. Voice AI company rocks into music territory!
6. 🚪💼 Three weeks after acquiring Windsurf, Cognition offers staff the exit door. Quick acquisition leads to quick departures!

What if your story was your sales strategy?

This week, I sat down with Maury Rogow, Hollywood producer and branding mastermind behind $2.5B in brand growth.

We talked about why the brands that feel effortless aren’t the loudest — they’re just the most aligned.

Google opens NotebookLM AI note-taking app to users 13+

⚡ You’re Smart. Strategic. Intentional. So Let’s Be Real:
Where Are You Holding Back— On Purpose?

Every smart operator has something they’re intentionally avoiding.

Not because they’re lazy — but because it’s risky, unknown, or just... a bit *uncomfortable*.

So what’s that thing for you?
That one decision, move, or experiment you know would push you forward — but you’ve been choosing not to do it.

Tell us — and we’ll share a tactical POV or tool to help you rethink it.
No fluff. Just momentum.

🛍️ Need a little dopamine hit? We got you.

Welcome to our brand new digital shopping corner — the place where scroll meets soul. Whether you’re here to procrastinate productively or just hunt down something unexpectedly awesome, you’re in the right place.

👉 Join our weekly AI tools building sessions – you can learn how to build and design for a fraction of the time: marketing tools, lead generation tools, and more.

👉 Sign up for the "Create How-To Guides with AI" tool here – fast, visual guides built with AI to make onboarding and documentation a breeze.

Go on, click it. It’s guilt-free indulgence (and might even be tax-deductible... kidding. Sorta.)

🎬 How to Use AI Agents to Build High-Converting Video Ads

Build once. Scale everywhere. In just minutes.

What if: Instead of spending weeks scripting, editing, and testing ads… you could build 10 variations of a high-converting video in 15 minutes — with agents handling the voiceover, visuals, call-to-action, and even the hook testing?

That’s exactly what we’ll show you this week — live.

Why This Changes Everything for Founders 🧠

• 🎥 Fast-Track Ad Creation: From idea to launch-ready video in under an hour — no team needed.
• 🔁 Test Like a Pro: Run A/B tests on 10 hooks, CTAs, or visuals — while you sleep.
• 💰 Better ROAS, Less Guessing: Let agents learn what your audience *actually* responds to.
• 📦 Works for Any Product: Digital or physical — agents adapt to your niche & brand voice.

📅 Join the Live Session w code Supercharged

We’ll walk you step-by-step through building your own ad-generating agent — no code needed.

👉 Click here to save your seat for the webinar

You’ll leave with a working prototype, free templates, and 5x more speed in your next ad campaign.

🎁 Bonus for Attendees:

• ✅ AI Agent Template for Ad Generation
• ✅ "Top 10 Hooks That Convert" Prompt Pack
• ✅ Entry to win a Free AI Strategy Audit