Upcoming Events

🌁 SF Bay Area

🗓️ Hungry for even more AI events? Check out SF IRL, MLOps SF, or Cerebral Valley’s spreadsheet!

🗽 New York

❗️The New York team is growing! We are seeking passionate volunteers to join us supporting event planning, sales and sponsorships, and marketing. Apply to Join the Team!

❝

The Taiwanese Asian Profession group of NY (TAP) is hosting an informal dinner revolving around AI and Taiwan. They are looking for those interested in the topic to join. Space is limited to around 10 attendees.

🏛️ DC

🎲 Las Vegas

🗻 Denver

🌴 Miami

🇮🇳 Delhi

🇮🇳 Bengaluru

🇮🇳 Hyderabad

Are We At The Tipping Point?

The Gen AI Collective’s record isn’t broken - we checked. And yet it sounds like it, as this past week has been (you guessed it): another milestone in AI history. Yet, across all 21 international chapters of the collective, we’ve been dancing to the music! Here’s what’s been playing out loud on our timelines, on both the “for you” and the “following” tabs: OpenAI’s GPT-4.5 launch, Anthropic’s release of Claude 3.7 Sonnet, and Manus’ First Autonomous Agent Launch.

Three major announcements. Strap in, it’s time to delve (Sorry, Paul Graham).

OpenAI’s GPT-4.5 Launch

❝

The News: OpenAI just launched GPT-4.5 (code-named Orion), the company’s most expansive model to date—utilizing unsupervised learning instead of traditional reasoning to achieve broader world knowledge and stronger emotional intelligence.

Details:

  • OpenAI claims GPT-4.5 offers a more natural conversational experience, better understanding of human intent, and heightened emotional intelligence.

  • The model hallucinates less and provides more accurate responses, making it useful for professional tasks, creative work, and everyday questions.

  • While it isn’t a major upgrade in math or science, it does outperform o3-mini and o1 on SWE-Lancer, OpenAI’s new freelance benchmark.

  • Pro users and developers with paid plans can access GPT-4.5 immediately, while Plus and Team users will gain access next week.

  • The API pricing is surprisingly high at $75/$150 per million input/output tokens, compared to GPT-4o’s $2.50/$10 rate.

❝

While GPT-4.5’s high pricing and minor upgrades might leave some users underwhelmed, it feels more like a “vibe upgrade” than a true technical leap. This could be the last iteration of non-reasoning AI models before OpenAI pivots toward a more hybrid approach. At the same time, OpenAI is shifting focus to $20K/month AI agents, moving beyond chatbots to specialized, autonomous AI workers. If GPT-4.5 refines the experience, these agents reshape the workforce—blurring the line between software and employee.

Claude 3.7 Sonnet Hybrid Reasoning

❝

News: Claude 3.7 Sonnet is a new Anthropic AI model that merges quick responses with deeper, step-by-step analysis. It launched with the same pricing as earlier Claude versions and is accessible via the Anthropic API, Amazon Bedrock, and Google Cloud's Vertex AI.

Details:

  • Offers two modes: Standard (updated from Claude 3.5) and Extended Thinking for multi-step reasoning.

  • Supports up to 128K tokens (beta), achieving 70.3% on SWE-bench Verified in Standard mode.

  • Maintains previous pricing of $3 per million input tokens and $15 for output, including thinking tokens.

  • Includes Claude Code for agentic coding, allowing automated debugging and refactoring.

❝

Claude 3.7 Sonnet isn’t just another AI upgrade—it’s a coding accelerator. With top-tier benchmarks and deep reasoning, it’s supercharging IDEs like Cursor, Windsurfer, and Replit, making AI-driven coding, debugging, and refactoring seamless. This means faster development cycles, fewer bottlenecks, and AI copilots that actually execute—not just suggest. Anthropic isn’t just competing; it’s reshaping how developers build software at scale.

'Fully autonomous’ Manus AI agent

❝

The News: China is experiencing its second DeepSeek moment, and this time, it’s called "Manus Mania." Manus, the world’s first fully autonomous AI agent, is making waves across the industry. At the time of writing, its Discord waitlist has surpassed 150,000 users eager to gain access.

Details:

  • Demos show that Manus can handle 50 browsing windows simultaneously, with tremendous performance.

  • Dubbed as ‘Claude MCP^1000 + Deep Research + Operator’ the benchmarks are some of the best we’ve ever seen.

  • Demonstrations show Manus handling tasks like resume screening and property research using its own independent computing environment.

  • Manus showcases skills like web browsing, coding, and visual content creation, all using it’s own independent computing environment.

  • It outperformed top general-purpose assistants like ChatGPT and Gemini on the GAIA benchmark for AI performance.

  • Manus is currently invite-only, with plans to open-source its models later this year.

❝

Just one day after OpenAI announced its pricing point for supercharged AI agents, Manus revealed an open-source alternative—a new champion for Open(space)AI. This marks the second major open-source disruption from China in under 90 days, signaling a shift toward democratized AI technology. U.S.-based companies may now face increasing pressure to justify high pricing models as global competitors introduce powerful, freely accessible alternatives.

Partner Spotlight

If voice is part of your next big idea, we might have found you something amazing.

The generative AI space has been moving fast, and one of the most common questions we get from founders is: What tools can I use to get started without breaking the bank? The reality is, most early-stage startups don’t have the resources to invest heavily in production from day one.

Our friends at Murf AI just opened up their API to everyone, and they’re launching a startup incubator program to help early-stage founders bring their ideas to life.

For those unfamiliar: Murf AI helps you create hyper-realistic voiceovers in 20+ languages. You can make voices say certain words exactly the way you would, switch between languages seamlessly, and add human-like nuances for a natural feel.

Now that that’s out of the way, here’s what you need to know about their API and startup incubator program:

  • 50 million free characters for AI voice generation (valid for 3 months)

  • 99.38% pronunciation accuracy—among the best in the industry

  • In blind tests, Murf AI’s voices won 8 out of 10 times for naturalness

  • The same voices can switch between languages with no loss in quality

If you’re building something that needs high-quality voice, this is worth checking out.

Apply for the Murf AI Startup ProgramExplore their API

Events Spotlight

Picture courtesy of Twama Nambili.

🌆 Seattle: ConvergeAI Summit

Seattle’s AI community gathered at ConvergeAI Summit for a full day of deep discussions, startup showcases, and thought-provoking panels. The event kicked off with a fireside chat featuring Matt F., who shared his journey from big tech to AI entrepreneurship. Later, Trent Gillespie delivered a keynote on leveraging AI to serve underserved markets, sparking conversations on accessibility and impact.

Attendees explored AI's evolving landscape through speed startup demos and pitches from founders like Melissa Hutchins, Ken Aragon, and Caleb John. A YoungTech Seattle panel rounded out the night, offering insights on AI’s next generation. Lively networking, a locally catered meal, and standout pitches made this summit a highlight for Seattle’s AI builders.

Picture courtesy of Ali G.

🍁 Toronto: UofT GenAI Chapter Launch

The UofT GenAI Collective launched with a packed house of AI researchers, founders, and industry leaders, cementing Toronto’s role as a hub for AI collaboration. Breakout discussions tackled real-world AI challenges, from business applications to cutting-edge research, offering attendees a chance to dive deep into emerging trends.

With support from Rootly, the event fostered meaningful connections between academia and industry, setting the stage for future partnerships. A post-event gathering at Pogue Mahone Pub kept the conversations flowing, proving that AI insights—and friendships—extend well beyond the main event.

Picture courtesy of Rochi Cairo Presepi.

🌁 San Francisco: Rho Private Dinner

A select group of AI founders gathered for a private dinner hosted by GenAI Collective and Rho, where candid conversations replaced formal pitches. Discussions covered the toughest challenges in scaling agentic AI—from fundraising and enterprise adoption to product testing and brand positioning.

Dubbed “Founder’s Therapy,” the evening created a space for builders to openly share insights and strategies. Over a chef-prepared meal, attendees explored the realities of launching AI-first companies in a rapidly evolving space. With an engaged group of founders and thought leaders, this invite-only gathering set the tone for deeper collaborations ahead.

Picture courtesy of Ashwini Iyer.

🌁 San Francisco: Beta University – Shaping AI Infrastructure

At Beta University, AI innovators came together to challenge assumptions and rethink infrastructure for the next wave of AI. A fireside chat featuring Nancy Zheng and Sakshi Pratap unpacked the technical and strategic hurdles of building AI-first companies, while an all-female founder panel highlighted breakthroughs in AI infrastructure.

The event drew a mix of startup builders and enterprise leaders, all eager to explore new approaches to foundation models, edge AI, and AI-powered hardware. Conversations continued well past the final session, proving that in AI, the best ideas often start with the right room of people.

Join the Community!

We are a volunteer, non-profit organization – all proceeds solely fund future efforts for the benefit of this incredible community!

🤝 Partner With Us

About Noah Frank

Noah is the co-founder of Aurix and has spent his career both working at startups and advising global leaders on innovation strategy. His work and body of research focus on AI policy, anticipatory governance, and effective decision-making. When not working to make emerging tech work for all, you can find him making music with his band. 🎸

About AJ Green

AJ is the Founder & CEO of AI Advantage Agency, WCYP Chairman, and Chapter Lead for GenAI Collective Portland. Former athlete turned tech entrepreneur, AJ is a builder and superconnector on a mission to make AI the great equalizer—helping startups scale and turning disruption into opportunity.

About Aqeel Ali

Aqeel co-leads the newsletter for the Gen AI Collective. He’s building Retentech AI, which helps ad supported websites boost revenue by retaining site visitors for longer. When not immersed in voice chats with ChatGPT, startups, or making untimely jokes, Aqeel writes! 🎨

Keep Reading