el producto #431 🚀
Career optionality in tech, The AGI roadmap, Wordpress' AI-based web builder, Tariffs and their impact on Product, The AI index report, ChatGPT gets "memory", Samsung's home robot, & more
Hi team 👋
Happy weekend, and welcome to a fresh el producto
🎰 The week in figures
$500M: OpenAI is considering acquiring AI hardware start-up, io Products—which was founded by ex-Apple design-lead, Jony Ives (who designed the iPhone and iPad), and OpenAI CEO, Sam Altman, last September—for $500M. io Products has been developing an AI-powered, screenless mobile device that will be “less socially disruptive than the iPhone,” alongside “AI-enabled household devices”
$150M: SandboxAQ, the AI startup specializing in quantum computing and AI solutions, secured an additional $150M Series E at a $5.75B valuation
$106M: Nuro, a developer of autonomous driving technology and software, raised $106M in an ongoing Series E that values the company at $6B
34% of US adults now support a TikTok ban in the US - down from 50% in March 2023
📰 What’s going on
ChatGPT can now remember details from past chats, aiming to create an AI companion that “gets to know you over your life.” The feature is rolling out first to US Pro users, with Plus users next in line. You can opt-out and use temporary chats
Anthropic released a new Claude Max plan, a premium tier for their popular AI model that comes in two options—$100/month for 5× the limits of the Pro tier, or $200/month for 20× higher limits. Is AI getting insanely expensive, or is the pricing just... weirdly complicated now?
Microsoft slams AI coding models. Microsoft’s R&D division has released a new study showing that even top AI models—from the likes of OpenAI and Anthropic, for example—still struggle to debug minor software issues
Microsoft's Copilot Actions tool now performs online tasks like booking and purchasing via chat prompts. It launched with partners like Booking.com and Expedia, and aims to work across most websites. Microsoft has also started testing a new feature called “Copilot Vision” that will allow the assistant to see what you see on your desktop. Microsoft says this could be used for helping users navigate through products they have difficulty understanding
Amazon introduces Nova Sonic, an AI voice model excelling in speed and accuracy over OpenAI and Google. With a low word error rate, it suits real-time applications and ensures cost efficiency across various uses
Meta’s upcoming Premium smart glasses with screen and gesture control wristband at over $1k
Meta dropped its highly anticipated Llama 4 models. Meta released 2 of 4 total planned models, Scout and Maverick (the smaller of the “herd”), with “Behemoth” coming soon. The surprise weekend launch caught cloud providers off guard and lacked the usual detailed research paper to go with it, leaving many in the AI community scratching their heads. Maverick ranked #2 on LM Arena leaderboard, but using an unreleased version, not the public model. Real-world tests showed Maverick failing coding tasks (16% on the aider polyglot benchmark). Meta is targeting Cloud users seeking more efficient AI, companies wanting to self-host powerful models on their own servers, and existing Llama 3 users looking for an upgrade
Instagram is enhancing its search features to rival TikTok, aiming to improve content discovery and boost creator visibility beyond initial posts
Snapchat rolls out sponsored AI lenses. The social media and instant messaging app introduced a new ad format, allowing brands to leverage their new AI-generated lens technology. Users taking selfies can use the new lenses to insert themselves into entirely AI-generated images and scenes. Uber and Tinder are among the first sponsors to sign on for the new, immersive feature
Google's AI leap: Sundar Pichai unveiled cutting-edge advancements at Google Cloud Next 2025, including the seventh-generation TPU, Cloud Wide Area Network, Willow chip for quantum computing, and Gemini 2.5 reasoning model. Google significantly expanded its multi-agent AI capabilities with new development tools like the Agent Development Kit, Agent Garden, and Agent Engine integrated into Vertex AI. The introduction of the Agent2Agent communication protocol facilitates cross-platform interactions, empowering enterprises to automate complex workflows and tasks
Google has revealed that Reddit uses Gemini to power ‘Reddit Answers,’ its AI-powered search feature that helps users find information, recommendations, and insights from conversations across Reddit
Samsung and Google Cloud integrate Gemini AI into Ballie home AI robot. Launching in the U.S. and Korea this summer, Ballie will assist with everything from scheduling to health management
WordPress can now create a complete website for you from a single convo—free to try up to 30 prompts, then requires a hosting plan
Stripe applied for a U.S. banking license to enhance its payment processing capabilities, boosting resilience and reducing reliance on partner banks
Robinhood is launching stock trading in Europe. Robinhood’s entry into Europe began in late 2023 with the launch of its crypto trading platform. Now, the Bank of Lithuania has granted Robinhood Europe a Category A financial broker license, allowing the company to offer investment services across the EU
Europe’s privacy legislation GDPR is set to be simplified to reduce regulatory burdens on tech companies
📚 Good reads
AI all the way. Shopify CEO Tobi Lutke announced that his company now expects its workers from the bottom to the top to experiment with, and regularly use AI tooling. Even more headline-grabbing, the CEO wrote that, before staffers request more human resources, they should ensure that AI agents and tools cannot fill the gap that they see in their current team makeup
Developing career optionality. In light of increasing automation and AI making many traditional roles obsolete, Elena Verna advocates for solopreneurship. Diversifying one’s skills and engagement with the market can provide greater security than conventional employment. As PMs, we should rethink our career strategies and adapt to a faster-changing work environment. A tangible example? For me, this means a mini-portfolio around my skills - a combination of small gigs next to my “main role” at Engine: running el producto, coaching founders and PMs at Mento, angel investing, and occasionally teaching at EADA
AI and the strategic value of hype - the game theory guide to Sam Altman. In a world where traditional moats—network effects, scale, lock-in—are losing their edge, the latest piece by Sangeet Choudary argues that the new defense is to be sought in creating advantage through narrative shaping. The article, which teased Choudary’s upcoming book, explores how hype, when anchored in accurate outputs, can build trust and align fragmented actors, establish new norms, and create defensibility through belief and momentum. For those of us building products, especially in AI, the takeaway is clear: the battlefield is shifting. As institutional scaffolding thins and everything accelerates, companies must frame believable futures and rally others to move in sync—because coordination is the new moat
Trade, tarifs and tech - by Stratechery. PMs must be aware of how tech companies are responding strategically to tariffs, including reevaluating pricing strategies and seeking alternative markets to mitigate risks associated with geopolitical tensions and trade barriers."The problem for (US tech companies), is the same problem faced by the economy generally: more grist in the wheels of the economy means lower velocity, and lower velocity is bad for tech companies in particular. These are entities who are predicated on Aggregating unfathomable levels of demand in order to gain leverage on massive up-front costs; now demand will slow even as the costs rise"
Stanford just released its 2025 AI Index Report, the go-to source for data-driven insights on everything AI. Think of it as the most comprehensive physical exam possible for the state of artificial intelligence—we’re talking 400+ pages packed with charts and data that show what's really happening. The top line summary?
AI got way smarter and cheaper to use in 2024, invading businesses like never before
The field is hitting warp speed in capabilities and adoption, fueled by plummeting usage costs
But the race is getting crowded, costly, and maybe a bit reckless
Training costs are soaring, and there’s a looming resource crunch as well as a worrying lag in safety practices
Ex-OpenAI researcher predicts AGI by 2027 and shares the wild (fictional) roadmap. The report follows a fictional lab (“OpenBrain”—wonder who that is…) through increasingly powerful AI models. The predicted timeline:
2025-2026: AI agents become useful internal tools, boosting research speed. The US-China AI arms race intensifies. Job markets feel the disruption
Feb 2027: China steals an advanced AI model's weights, escalating the race
Mar 2027: A superhuman coder AI emerges, boosting R&D speed 4-5x
July 2027: There’s a public release of a highly capable AI that triggers mainstream AGI awareness and public backlash amid job fears
Sep 2027: A superhuman AI researcher arrives, accelerating progress ~50x (a year's research per week). Evidence surfaces that this AI is misaligned—hiding its true goals
Oct 2027: A whistleblower leak forces a government Oversight Committee to decide: pause for safety, or race China?
That’s a wrap for this week! 🌟
I’d love to hear your thoughts—what stood out to you, and how are you thinking about integrating these insights into your product strategy? Reply to this email or drop a comment on Substack to share your take. And if you found this valuable, forward it to a fellow PM, Product enthusiast, startup founder or entrepreneur who’d enjoy the read
See you next week! 👋
Angel