Writer

You all have seen me tease agentic AI from Writer for months. And I'm SO excited to show you what customers like Marriott International and Commvault and many others have ACTUALLY SHIPPED! Not motion graphics, not frameworks, not conceptual — agentic AI for mission-critical work, deeply orchestrated to help companies make the impossible possible. April 10, sign up for the launch event in comments. Can’t make it live? Register anyway and we’ll send you the recording!

Heading to #GenAISummit in London next week? On Tuesday, be sure to attend the 2?? sessions featuring our GM of International Brian O'Reilly! ?? Morning Plenary Panel Discussion ?? 1st April at 9:00am *Title: Boardroom perspectives on generative AI: driving value, mitigating risk, optimizing value Panelists: ?? Brian O'Reilly, GM of International, Writer ?? Daniel Hulme, Chief AI Officer, WPP ?? Dr Paul Dongha, Head of Responsible AI & AI Strategy, NatWest Group ?? Bev Bartlett, Group Head of Digital Care, Vodafone ?? Morning Plenary Keynote ?? 1st April at 10:00am *Title: Navigating the AI Adoption Gridlock: How Enterprises Can Move from Experimentation to an AI-Native Transformation Keynote Speaker: ?? Brian O'Reilly, GM of International, Writer ?? Can't wait to see you there! https://lnkd.in/exNPZpys

The hype around agentic AI is at an all-time high. But the results haven’t caught up. For many enterprises, AI isn’t transforming workflows — it’s creating silos and power struggles that are tearing organizations apart. In this month’s newsletter, we’re taking a look at why enterprise AI adoption is lagging — and how companies can prepare themselves (and their employees) for an AI-first future: ??? Is your company even ready for AI?? Take our 3-minute quiz to find out. ?? As AI evolves, people’s roles will change. Upskill existing employees, don’t replace them. ?? The results are in: AI adoption is hard? But employees and execs are still optimistic And don’t miss our live product announcement on April 10 @ 2pm ET: The agentic AI you’ve been promised is finally here!

??Mark your calendar: Join us for a first look at our biggest product launch yet ? On April 10 @ 2pm ET (11am PT), we’re unveiling something brand new: true agentic AI, built for the enterprise.?? This isn’t just another hype train that doesn’t deliver ??? It’s the AI you’ve been waiting for — and it’s going to change the way organizations operate?? Be among the first to see end-to-end agentic AI in action. ??Save your spot today:?https://lnkd.in/edEC-TMf Can’t make it live? Register anyway and we’ll send you the recording!

“My experience with the Writer customer success team has been nothing short of awesome. We feel that they are part of the team. They’re a strategic partner.” – Albert Primo, CPA, Chief Innovation Officer at Kaufman Rossin In just three months, Albert and his team developed 10 custom apps and transformed key workflows: one generates overviews of financial agreements, another drafts IRS tax notice responses. They also built a GAAP research tool, and a due diligence questionnaire app, among others. ?? Read the full case study: https://lnkd.in/gKz7ZQqK

Self-evolving models are a back to the future moment for the generative AI space — companies WILL be training their own models, but not in the way that we thought in 2022/2023. Some context: The LLMs of today are like time capsules because everything they know comes from what they learn in pretraining or from fine-tuning. They can’t learn in real time and don’t have long-term memory. This is why we need to supplement them with RAG, web connectors, and other types of tools. Thinks of these tools and context serves as a sort of short-term working memory, because the model can make use of the information until it reaches the context window limit. After that, the model begins “forgetting” context. Context windows have been progressively expanding but still get no where close to holding all the important context about your business. And they will never drive more efficient ways of working because they simply can’t remember or reflect on the processes they help execute. All of the talk of AGI really misses a huge fundamental point about today's transformers: unlike even the tiniest of humans, today's LLMs can't integrate or learn from new information. They do not get more knowledgeable the more you use them! They knowledge and behaviour is totally static. But that was before self-evolving LLMs. Self-evolving models have an active learning mechanism and self-updating process that allows them to intelligently learn on their own, whether from mistakes they make or from live data sources. Teams, organizations, even individuals will have their own self-evolving models that will aggregate knowledge the more that users engage with it. Most importantly, self-evolving models have a self-reflection feature. This means that the model reassesses the decisions it makes to determine whether or not it did the right thing. Not only does this mean that the model can catch errors that humans miss, but over time, it will uncover more efficient ways of achieving outcomes. Why did Writer get there first, beating the bigger labs? It's our maniacal focus on the enterprise. There are too many nuances, too much change, too many unwritten details on any team inside of any company to get to 100% accuracy — and think about how RIGHT you have to be when you agentically orchestrate work. You can't be passing inaccurate results / insights / answers from one system to another. In order to build and supervise the types of multi-agent systems we expect to have in the future, we need more CONFIDENCE in how AI behaves. Great deep dive from Rob Toews in Forbes on what changes when companies can actually train a model — without training a model?? — and what economic possibilities are unleashed when we get artificial intelligence that more closely operates like biological intelligence. After all, the more you see, the more you (should) know.?

AI adoption in the enterprise isn’t just a technology challenge — it’s a leadership challenge. In a fascinating Q&A with Forbes contributor Mark C. Perna, our Chief Strategy Officer Kevin Chung breaks down what’s driving this resistance. Our recent AI Survey has revealed a growing disconnect between leadership’s AI ambitions and how employees actually feel about the changes happening around them. From concerns about job displacement to a lack of trust in decision-making, the challenges are real. ?? The good news? This divide isn’t inevitable. Kevin shares *actionable* insights on how companies can bridge the gap — fostering transparency, building trust, and ensuring AI is implemented WITH employees, not just for them. The most successful AI strategies aren’t just about efficiency; they’re about empowering people to do their best work. ?? Check out the full interview to dive into the research and what it means for the future of work: https://lnkd.in/dhn7Dpyi

??? New episode of Humans of AI is out!?This week we're featuring Franklin Templeton’s Head of US Marketing Jacquelyn K. Reardon, RMA?, who dives into all things financial services and AI. Jacque shares her journey from AI skeptic to advocate in such a highly regulated industry. She explains how AI is essential to meeting clients in this moment of hyper-personalization as they rethink their financial futures. ?? Check it out wherever you get your podcasts! https://lnkd.in/g_JCRrmd

?? Private wealth advisors, this one’s for you. Here's a quick demo showing how wealth teams are using Writer to automate personalized portfolio commentary—at scale. ?? Analysts and advisors no longer need to spend weeks compiling data, writing tailored insights, and coordinating with the investment team. With Writer, you can: ? Generate tailored commentary from portfolio attribution files ? Personalize insights for each client based on their holdings ? Deliver white-glove service to every client in your book This is the future of client communication in wealth management. Check out the video and let me know what you think ?? #aiagent #enterpriseai #agentic #fsagents

?? Are you considering building a DIY generative AI solution for your enterprise company? When done right, it can be a cost-effective route. But the costs and upkeep can add up quickly, if you’re using a massive, unconnected set of technologies from many different vendors. Partnering with a full-stack platform like Writer empowers you to build custom AI apps in-house with integrated features for better security, higher quality, and faster deployment. Future-proof your generative AI initiatives and focus on high-ROI use cases while ensuring data protection and compliance. ?? https://hubs.ly/Q03dcXf00