Zylon by PrivateGPT

Next week I will be presenting at #18ENISE organized by INCIBE - Instituto Nacional de Ciberseguridad the result of the collaboration between Zylon by PrivateGPT & Occentus Network: "Zylon in a Box, the ready-to-go Private AI for the Enterprise" 🚀 With "Zylon in a Box" we have made even easier to deploy Zylon by leveraging Occentus' secure and cost-efficient infrastructure with 24/7 support certified by NIS2 and ISO 20000, 27001, 27017, 27018 & 22301. Now companies can forget about the setup and maintenance associated with running AI within their systems and let Occentus do it for them, while keeping 100% of Zylon's AI-powered features. If you want to know more, visit our stand (number 157) or join me at the Speaker Corner 1 on October 22 17:20, where I will be providing a talk on "How to adopt Private AI while keeping data sovereignty". See you there!

Since we started our Go-To-Market strategy for Zylon by PrivateGPT by early August, the conversion rate from the initial demo session to the following project scoping meeting has been nearly 100%. Every enterprise wanting to adopt Private AI has immediately recognized the unique value provided by Zylon to keep data sovereignty within their infrastructure. Our all-in-one collaborative AI workspace can be deployed within hours (both private cloud or bare metal) after defining the infrastructure requirements to run it along with the organization's IT team. But we know that not all organizations have the capacity or expertise to set up and maintain such infrastructure. That's why we've been providing customized support to help them get started so far. However, we've found an even better solution: partnering with leading cybersecurity hardware providers to simplify the setup process. Today, I'm thrilled to announce our partnership with Occentus Network SL, a top-tier security infrastructure provider with 24/7 support certified by NIS2, ISO 20000, 27001, 27017, 27018 & 22301. Together, we're launching "Zylon in a Box" - the simplest way to adopt Private AI within your organization without worrying about infrastructure details. Stay tuned or contact me directly to learn more about this game-changing partnership!

Good news everyone! We have been selected by El Referente to be part of the list "Top 100 Startups 2024" 🎉 After only 1 year of adventure making possible for enterprises to adopt Private AI while keeping data sovereignty, we are really grateful to see that the work we are doing at Zylon is so well received by the entrepreneurship ecosystem.

"Enjoy the journey to shape the future" is one of the core cultural values at Zylon by PrivateGPT, something that we nourish on a daily basis. For a fully remote company like us, off-sites are key events to reinforce the social ties that create the space for innovation and disruptive ideas fueling our product. This time we have enjoyed 3 amazing days in Alicante (Spain) full of enlighitng conversations, laughs and everlasting memories. I'm a really lucky to be surrounded by such a talented and creative team, but specially by great human beings in which I can trust the future of Zylon. Let's make history together helping organizations to adopt private AI while keeping data sovereignty!

We are excited to be part of the finalists for the ISS A/S Startups Awards in the category "Best Technology Initiative to improve the Workplace" alongside Payflow (YC S21), Imperia SCM and Zepo. Thanks to elEconomista for spreading the word! https://lnkd.in/dvR3ijWz

Exciting news! Join me next Thursday as I dive into the critical topic of privacy risks associated with AI adoption and how you can leverage PrivateGPT (https://privategpt.dev) to overcome them in your organization 🔒 Thanks to Zilliz and Milvus for inviting me to share my expertise with their community 😊 Register here: https://lnkd.in/dgzxprEh

Kicking off the weekend with exciting news! 🚀 We're thrilled to welcome Cristina Traba Deza as our new Senior Product Designer, bringing her talent and passion to our team! With a strong foundation in Architecture, she forged her Product Design career at innovative unicorns like Cabify and Revolut, where she played a pivotal role in creating high-impact features that reached millions of users across mobile and web platforms. Her holistic approach spans from customer research to pixel-perfect design, ensuring seamless user experiences. Super happy to have Cris on board as we continue to push the boundaries of product design to deliver the truly private AI workspace 👏

I'm very aligned with Aidan Gomez, Cohere CEO, on the challenges and future of enterprise AI adoption. Here are some common design decisions in Cohere and Zylon: - On-prem deployment is key for privacy: SaaS solutions may be suitable for consumers, but they're not viable for enterprises that require strict data control. - Multilingual support is essential: AI tools must be able to process and understand multiple languages to effectively serve diverse users and documents. - Explicit explainability is crucial for accuracy: Users must be able to directly verify answers through inline references (see how we do it in Zylon's screenshot below). - From knowledge to automation: Today, AI enables companies to leverage their knowledge like never before. Tomorrow, enterprises will integrate private AI with their tools to automate entire workflows. Zylon by PrivateGPT takes a different approach than Cohere, offering a ready-to-use, customizable AI workspace that enables companies to adopt AI without initial development efforts. In contrast, Cohere provides technical foundations for organizations with the resources and expertise to build custom end-user solutions. Both approaches are necessary in the enterprise market. Source: QuantumBlack, AI by McKinsey https://lnkd.in/dUyZTntn

I found another interesting take from the CNBC article on J.P. Morgan private AI adoption plan. Teresa Heitsenrether is 100% correct in breaking it down into three sequential stages. 1. "The first is simply making the models available to workers" -> while all want to rush into complex use cases, there is a whole amount of value in prompting-based interactions, especially when a private setup allows inputting private information in the prompt. Covers 20-30% of common use cases. 2. "The second involves adding proprietary JPMorgan data to help boost employee productivity" -> contextualized AI interactions are where things get interesting: querying documents, gaining insights from sparse data sources, and innovating on top of previous experiences. Covers 40-50% of common use cases. 3. "The third is a larger leap [...] when generative AI is powerful enough to operate as autonomous agents that perform complex multistep tasks." -> Call it agents or workflows; the future is clearly on bringing custom processes into private AI platforms to get them semi-to fully automated. It's in the high-investment, high-value part of the matrix, covering the remaining use cases. We follow the exact same approach with our customers, as we've validated that this method yields a faster return on investment. Original article: https://lnkd.in/dW97qgV5

Large organizations in regulated industries are leading the way, but the problem they are solving affects all businesses: AI privacy. "The bank is giving employees what is essentially OpenAI’s ChatGPT in a JPMorgan-approved wrapper more than a year after it restricted employees from using ChatGPT. [...] We’ve implemented it in a way that we can leverage the model while still keeping our data protected." We created Zylon by PrivateGPT to help organizations of all sizes deploy secure and compliant AI workspaces quickly and easily. And it's working - we're being contacted daily by financial institutions, healthcare organizations, insurance companies, public administration, and other forward-thinking businesses that need a reliable solution. The future of enterprise AI is on-prem, not in the cloud. See you there.