Databricks’ Post

Great insights from the State of Data + AI report by Databricks. Discover how companies are prioritizing AI and open-source models. Read the full report. #DataAI #AIInsights #OpenSourceAI #LLMs #TechTrends #Innovation

In the State of Data + AI, we analyzed the usage of Meta Llama and Mistral, the two biggest players of open source LLMs. Key findings: - 77% choose models with 13B parameters or fewer - 76% of companies using LLMs choose open source, often alongside proprietary models. Find out how this translates to company priorities in our new report: https://bit.ly/4buAgDH

Slides from IIPP’s research showcase on Algorithmic Rents are now available! Discover the research of Mariana Mazzucato Tim O'Reilly, Ilan Strauss, & Rufus Rock that shows how BigTech extracts rents from various actors in their ecosystems & what we can we do to better govern AI. Link here: https://lnkd.in/eZbts823

ICYMI: last week Forbes shared their 2024 AI 50 List, where 6 Lightspeed portfolio companies were recognized as the most promising privately-held artificial intelligence companies: Abridge, Anduril Industries, Glean, Mistral AI, Pika, and Tome. We’re excited to see how this year unfolds as they continue to be at the forefront of generative AI: https://lnkd.in/eTKf9N4U

What is a successful use case of AI's LLM power to harness data? Quoting Brian Buchwald: We - at Edelman - have a robust database of surveys and quantitative research [on topics] associated with trust, such as ability, dependability, relevance, integrity and purpose. We understand how each of these actually contribute to this measure of trust, so that for any industry, I can show you how important it is to be seen as competent versus innovative, or have high integrity versus being purposeful. We have definitions for each of those pillars, which contribute in their own way up to an overall trust score. And then I have 3 million articles or pieces of content a day that I’m scoring on those pillars and overall trust. It’s one thing to tell [a brand] that the digital data is telling us their trust is going to go up or it’s going to go down. It’s very different for me to be able to use Archie and say specifically, “These articles are contributing to or detracting from your trust. These specific topics, these specific journalists, these specific publications are either aiding you or diminishing trust amongst your audiences.” Article via Ad Age https://lnkd.in/df6hQCjM

Watch now: 2024 Imperva Bad Bot Report, AI & more explored with Reinhart Hansen Nearly half of all global traffic now comes from bad bots - which is a shocking finding. Learn more in my great video interview with Imperva's Director of Technology in the CTO Office, Reinhart Hansen! Full article and video interview here: https://lnkd.in/guekqgjN

I've gotten to present this topic for a few organizations; it's pretty basic and covers: - What is generative AI - What's possible with generative AI today (with examples) - How can your organization harness AI today - What are some predictions of the near-term future with generative AI? Rotarians like it, manufacturing leaders like it, and even high schoolers sat through the whole thing. I'd love to see you there too.

Today, Google introduced the world to Gemini, a new AI model which was the first to outperform human experts on MMLU (Massive Multitask Language Understanding). Check out this demo. It is pretty mind blowing! Read the latest on building with Gemini: https://goo.gle/3NeU4Sq And join the Applied AI Summit on Dec 13th to learn more: https://lnkd.in/grDtE7Jc #itsaduck

Debate, discuss, test... What's the worst that could happen? Carolyn Dawkins, CMO for David Yurman discusses the tools, skills, and culture needed to be insights-driven, the opportunities and threats of generative AI, and how to blend data with creativity 🎨🧪 Watch the full session, recorded at Emarsys Power To The Marketer New York 2023: https://bit.ly/489nQiN #PTTMFestival2023 #Data

🙌 Get in the Mix...tral! IBM is continually growing our library of proprietary and 3rd party open source models in watsonx.ai with the inclusion of Mixtral-8x7B! Think of AI models as tools. There is no "one size fits all". Invest in a platform that gives you flexibility to test/train/deploy DIFFERENT models in a SINGLE platform. "IBM offers an optimized version of Mixtral-8x7B that, in internal testing, was able to increase throughput — or the amount of data that can be processed in a given time period — by 50 percent when compared to the regular model.1 This could potentially cut latency by 35-75 percent, depending on batch size — speeding time to insights. This is achieved through a process called quantization, which reduces model size and memory requirements for LLMs and, in turn, can speed up processing to help lower costs and energy consumption.