Writer

Office hours are now open! Save your spot for tomorrow's first session: https://hubs.ly/Q0343jMl0 Join CPG industry expert Kushala Silva (Amazon, P&G, Coca-Cola) and Johanna Murphy, AI Program Delivery Manager at Writer, for a deep dive into how generative AI can transform product detail page (PDP) workflows. Bring your questions! Learn how consumer goods companies can use #generativeAI to: 🔎 Improve product discoverability ⬆️ Increase conversions ⬇️ Reduce overhead 💰 Boost revenue Live on January 28 at 11am PST, 2pm EST. Recording will be available for those who can’t make it.

A tech sell-off in response to DeepSeek is absolutely insane. It’s like freaking out because some team in a garage also invented the combustion engine in 1860, before anyone actually had a car. Engines didn’t change the world. Henry Ford did, with his invention of the assembly line. The real insanity is how good the models are already and how little the transformation is at most companies. We've been beating this drum for two years: the vast majority of companies have not seen significant ROI on their generative AI investments. That’s what should be making the headlines! But it’s ALSO why people should be so bullish. Our full-stack approach combines state of the art models with the logic and data the models need to actually be useful. And I’ve seen what the Writer agentic capabilities are and the early results we’ve seen with our customers. It’s mindblowing. The productivity per employee in the US is going to 10x before 2030. Why the US? Because it’s the only place I’ve seen enterprise leaders obsessed with disrupting their way of doing things. Like execs working nights and weekends for the first time since their 20s. And that true disruption is more than just the LLM. It’s the will to do whatever it takes to rebuild AI-first. DeepSeek followed the path we’ve pioneered at Writer — high quality synthetic data and precision training. And the many engines to come? All of this is a good thing for the market. This world needs all sorts of engines. There are engines for lawnmowers, go-karts, motorcycles, airplanes, and rocket ships. The trick is knowing the right engine for the right job, ensuring what you're building around that engine is extremely high quality, and that it's going to solve the problem it's designed for. You need to know when you need a motorcycle and when you need a rocket ship. You need to make sure the user knows how to drive it. And you better be damn sure that engine isn't going to give out when a newer model hits the shelves. This just validates what we’ve been saying all along. The future belongs to those who can assemble the rocket. DeepSeek can make you bearish on OpenAI but no way can it make you bearish on the coming explosion of inference. It hasn’t even really started.

Davos 2025: What an incredible week, as our conversations now turn to reflections, ideas, and actions for the year. The theme at #WEF25 was Collaboration for the Intelligent Age, which speaks directly to Writer's human-centric approach to innovation and transformation. This theme was reflected in all things Writer — from our airport ads upon arrival, to May Habib and Kevin Chung's speaking sessions, to the events with partners including Slack and Boston Consulting Group (BCG), to the informal conversations that only Davos can provide. 🏔️ You can watch May's sessions and interviews on-demand to inform and inspire your generative AI journeys: ✅ The The Female Quotient panel: https://lnkd.in/gAnknwff ✅ World Economic Forum Spotlight session: https://lnkd.in/gmcyDvEM ✅ Axios fireside chat: https://lnkd.in/g9cxBAvq ✅ Interview with Fox Business Network: https://lnkd.in/gxvwHGC3 Special thanks to the customers, partners, prospects, and industry leaders we had the privilege of meeting last week. 🙏 We're excited to strengthen our collaboration as we work to build a human-centric future of AI.

“When I work with organizations around embracing emerging tech like AI, it's actually about building culture.” With a background influenced by both art and computer science, Chantal Forster (AI Strategy resident, Annenberg Foundation) finds that the key to helping organizations embrace generative AI isn't just about the technology itself. While tech fluency and strategy are important, she says the core of successful tech integration is about creating the right conditions for people to thrive and adapt. 🔗 Check out the full conversation on this episode of Humans of AI — Presented by Writer wherever you get your podcasts: https://ow.ly/rAFU50UNiJk

Get real ROI from your generative AI investments: Check out our new Big Book of use cases for retail and consumer goods. 📗 Updated for 2025, the guide includes: 💥 Real-world use cases — learn how custom Writer solutions can make the entire product lifecycle more efficient, from development to customer support 🚀 Customer success stories — Adore Me saves 35 hours per person every month. A global #CPG brand saves $500K annually 📈 Tailored solutions — find the generative AI use cases that will drive adoption and ROI across your organization Download your free guide at https://hubs.ly/Q033_zXY0

Lots of hype about DeepSeek and its R1 LLM released this week - but DeepSeek shouldn't be a surprise to anyone. In October, Writer released a top-benchmarked frontier model, Palmyra X 004, that cost $700,000 in GPUs to train, developed entirely with synthetic data. Back in July 2023 we published research on how to develop these models with fewer GPUs. It simply doesn't take billions of dollars to achieve state-of-the-art model performance. Here’s the link to our research: https://lnkd.in/dd5jg-Be It really comes down to incentives. Our incentive is for our customers to run these models on their own infrastructure and in their own systems. We're a company, not a research lab. And for China and DeepSeek, they were incentivized to find innovative workarounds given chip constraints. The biggest AI labs have zero incentive to prevent losing billions of dollars on training. They're playing a different game. If they had the same constraints and same goal, they would end up eventually at the same point.

Wrapping up a "priceless" week at #WEF25 with Mastercard Chief Marketing & Communications Officer Raja Rajamannar and May in Davos! 🏔️ Stay tuned for our event recap posts coming soon 👀

✅ Unmatched accuracy when generating content ✅ Compliance with global privacy and security standards ✅ Easy deployment and smooth scaling ✅ 67% lower costs compared to DIY solutions Our graph-based RAG, Writer Knowledge Graph, unblocks many of the biggest barriers to full-scale enterprise AI adoption. It’s easy to implement, simple to maintain, and fully transparent so you can always trust the results. Plus, it’s a fraction of the cost of DIY vector retrieval solutions. 

 Get started with Knowledge Graph in five steps — download the guide: https://hubs.ly/Q033R_Vk0

Amazing discussion with Writer, Boston Consulting Group (BCG), BCG X, NVIDIA, and other industry leaders at #WEF25! BCG and BCG X hosted a dynamic conversation for an exclusive group of execs at the historic Schatzalp Hotel. It was standing-room only at the hotel known as the "Magic Mountain" with its sweeping views of Davos! 🏔️ A world-class speaker lineup (including our own Kevin Chung) gave the opening remarks, which sparked a great back-and-forth interaction between Writer, BCG, BCG X, NVIDIA, Merantix Momentum, and all the attendees. Huge thanks to the featured speakers: 🔹 Jessica Apotheker, Global CMO, Boston Consulting Group (BCG) 🔹 Nicole Büttner, Founder & CEO, Merantix Momentum 🔹 Kevin Chung, Chief Strategy Officer, Writer 🔹 Sylvain Duranton, Global Leader, BCG X 🔹 Dr. Eva-Maria Hempe, Head of Healthcare & Life Sciences, EMEA, NVIDIA Special shoutout to Tapiwa Meda for the photos and support on social! 👏

🎤 Thanks to Axios and Managing Editor Alison Snyder for hosting a great one-on-one chat with May Habib in front of a packed room at #WEF25! Their conversation covered important AI themes echoed throughout the week, with May offering her unique Writer perspective based on real customer deployments and insights. See the full interview below 👇 🎙️ Excerpts from May's conversation with Alison: Alison: “With AI agents, how big is the gap between what they can currently do and their promise?”  May: "The gap is not about product or technology [we know how to solve these problems]. The biggest gap we see is business process." Alison: “Do you encounter resistance from workers using AI tools?” May: "It is rare, honestly. People are so hungry for more exciting tooling. I think they've been drastically underwhelmed by the quality of what they've seen. But they're hungry for good tooling." Alison: What should we be paying attention to that we're not? May:  "Learning to use AI in your day-to-day is like re-learning how to read and write. We are democratizing what it means to be a builder....We can come up with much more highly personalized experiences to get our work done."