🏭 Industrial IoT + AI / Generative AI 🤖: Actual Case Studies, Insights and Perspectives >>> 🚀🚀🚀 | May - June 2024
🔹Fabio Bottacci
Senior Business Advisor / Venture Partner | Industrial IoT + AI / ML > GenAI > Agentic AI Expert | Executive Director, Business Development | B2B Startups GTM Strategy & Execution / Funding | HBR Advisory Council Member
Hi everybody,
Just some stimuli for your brain during the so-desired summer readings... 🏖📖🏖📖🏖
Enjoy your holidays!
Fabio
#industrial #iot #iiot #ai #genai #generativeai #casestudies #insights #perspectives
------------------------------------------------------------
🔹Liked this newsletter? 🔹Want to see more? 🔹Subscribe!
Continental 𝐚𝐧𝐝 SoftServe 𝐃𝐞𝐥𝐢𝐯𝐞𝐫 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈-𝐏𝐨𝐰𝐞𝐫𝐞𝐝 𝐕𝐢𝐫𝐭𝐮𝐚𝐥 𝐅𝐚𝐜𝐭𝐨𝐫𝐲 𝐒𝐨𝐥𝐮𝐭𝐢𝐨𝐧𝐬 𝐰𝐢𝐭𝐡 𝐎𝐩𝐞𝐧𝐔𝐒𝐃: 𝘞𝘪𝘵𝘩 𝘢𝘶𝘵𝘰𝘮𝘰𝘵𝘪𝘷𝘦 𝘤𝘰𝘯𝘴𝘶𝘮𝘦𝘳𝘴 𝘪𝘯𝘤𝘳𝘦𝘢𝘴𝘪𝘯𝘨𝘭𝘺 𝘴𝘦𝘦𝘬𝘪𝘯𝘨 𝘮𝘰𝘳𝘦 𝘴𝘦𝘢𝘮𝘭𝘦𝘴𝘴, 𝘤𝘰𝘯𝘯𝘦𝘤𝘵𝘦𝘥 𝘥𝘳𝘪𝘷𝘪𝘯𝘨 𝘦𝘹𝘱𝘦𝘳𝘪𝘦𝘯𝘤𝘦𝘴, 𝘵𝘩𝘦 𝘪𝘯𝘥𝘶𝘴𝘵𝘳𝘺 𝘩𝘢𝘴 𝘪𝘯𝘤𝘳𝘦𝘢𝘴𝘦𝘥 𝘪𝘵𝘴 𝘧𝘰𝘤𝘶𝘴 𝘰𝘯 𝘤𝘰𝘯𝘯𝘦𝘤𝘵𝘪𝘷𝘪𝘵𝘺, 𝘢𝘥𝘷𝘢𝘯𝘤𝘦𝘥 𝘤𝘢𝘮𝘦𝘳𝘢 𝘴𝘺𝘴𝘵𝘦𝘮𝘴, 𝘢𝘯𝘥 𝘵𝘩𝘦 𝘪𝘯-𝘷𝘦𝘩𝘪𝘤𝘭𝘦 𝘦𝘹𝘱𝘦𝘳𝘪𝘦𝘯𝘤𝘦. 🚀🚀🚀 | Via NVIDIA Developer | May 1, 2024 | @VINCI_Digital
Continental , a leading German automotive technology company and innovator for automotive display solutions, is developing AI-powered virtual factory solutions to address these shifts and their impact on their 𝐦𝐚𝐧𝐮𝐟𝐚𝐜𝐭𝐮𝐫𝐢𝐧𝐠 𝐨𝐩𝐞𝐫𝐚𝐭𝐢𝐨𝐧𝐬 such as shorter development cycles and increased time-to-market pressures.
🔹 As a pioneer of sustainable and connected technologies in automotive, their solutions are using the latest simulation and 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 technologies to transform manufacturing operations, enhance team productivity, and accelerate innovation to meet changing consumer needs.
🔹In addition, Continental leveraged the expertise of SoftServe, an IT consulting and digital services provider and trusted NVIDIA partner, to add 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 𝐟𝐮𝐧𝐜𝐭𝐢𝐨𝐧𝐚𝐥𝐢𝐭𝐢𝐞𝐬 𝐭𝐨 𝐭𝐡𝐞𝐢𝐫 𝐯𝐢𝐫𝐭𝐮𝐚𝐥 𝐟𝐚𝐜𝐭𝐨𝐫𝐲 𝐬𝐨𝐥𝐮𝐭𝐢𝐨𝐧𝐬 𝐚𝐧𝐝 𝐟𝐮𝐫𝐭𝐡𝐞𝐫 𝐞𝐧𝐡𝐚𝐧𝐜𝐞 𝐭𝐡𝐞 𝐞𝐟𝐟𝐢𝐜𝐢𝐞𝐧𝐜𝐲 𝐨𝐟 𝐭𝐡𝐞𝐢𝐫 𝐞𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐢𝐧𝐠 𝐭𝐞𝐚𝐦𝐬.
Source: https://bit.ly/4drHcDC
----------------------------------------------
#iot #ai #genai #virtualfactory #automotive #manufacturing
𝐓𝐡𝐞 𝐍𝐞𝐱𝐭 𝐀𝐜𝐭 𝐨𝐟 𝐆𝐞𝐧𝐀𝐈: 𝐇𝐨𝐰 𝐭𝐨 𝐆𝐞𝐭 𝐀𝐈 ‘𝐀𝐠𝐞𝐧𝐭𝐬’ 𝐖𝐨𝐫𝐤𝐢𝐧𝐠 𝐋𝐢𝐤𝐞 𝐇𝐮𝐦𝐚𝐧𝐬: 𝘈𝘐 𝘢𝘨𝘦𝘯𝘵𝘴 𝘩𝘢𝘷𝘦 𝘨𝘦𝘯𝘦𝘳𝘢𝘵𝘦𝘥 𝘢 𝘭𝘰𝘵 𝘰𝘧 𝘩𝘺𝘱𝘦 𝘪𝘯 𝘵𝘩𝘦 𝘱𝘢𝘴𝘵 𝘺𝘦𝘢𝘳, 𝘣𝘶𝘵 𝘢 𝘯𝘦𝘸 𝘳𝘦𝘱𝘰𝘳𝘵 𝘰𝘶𝘵𝘭𝘪𝘯𝘦𝘴 𝘩𝘰𝘸 𝘵𝘩𝘦𝘺 𝘵𝘰 𝘣𝘦 𝘪𝘮𝘱𝘳𝘰𝘷𝘦𝘥 𝘵𝘰 𝘣𝘦𝘤𝘰𝘮𝘦 𝘮𝘰𝘳𝘦 𝘶𝘴𝘦𝘧𝘶𝘭 🚀🚀🚀 | Via Sifted / Backed by Financial Times | May 9, 2024 | @VINCI_Digital
For over a year now — a long time in the era of 𝐆𝐞𝐧𝐀𝐈 — 𝐀𝐈 “𝐚𝐠𝐞𝐧𝐭𝐬” have been the next big thing, right around the corner.
🔹Mainstream excitement around these tools — which promise to optimise powerful large models 𝐭𝐨 𝐜𝐚𝐫𝐫𝐲 𝐨𝐮𝐭 𝐚 𝐬𝐞𝐭 𝐨𝐟 𝐭𝐚𝐬𝐤𝐬 𝐨𝐫 𝐣𝐨𝐛𝐬 𝐨𝐧 𝐭𝐡𝐞𝐢𝐫 𝐨𝐰𝐧, rather than just answer queries from humans one by one — kicked off with the release of 𝐀𝐮𝐭𝐨𝐆𝐏𝐓 by Scotland-based developer Toran Bruce Richards in March 2023 to a fanfare of posts on AI Twitter.
🔹But, as with many developments in the world of 𝐆𝐞𝐧𝐀𝐈, the delivery of genuinely useful results or products didn’t swiftly follow, and it seemed like most people quickly forgot about making new things with tools like 𝐀𝐮𝐭𝐨𝐆𝐏𝐓.
🔹Now, a new report from Amsterdam-HQed software developer and investor Prosus Group suggests that these 𝐀𝐈 𝐚𝐠𝐞𝐧𝐭𝐬 are beginning to become useful in some business contexts, albeit with some fairly serious limitations.
Sources:
----------------------------------------------
#iot #ai #genai #llms #aiagents #agentops #autogpt
GrayMatter Robotics 𝐒𝐜𝐨𝐫𝐞𝐬 $𝟒𝟓𝐌 𝐟𝐨𝐫 𝐑𝐨𝐛𝐨𝐭𝐬 𝐭𝐡𝐚𝐭 𝐒𝐩𝐞𝐞𝐝-𝐮𝐩 𝐌𝐚𝐧𝐮𝐟𝐚𝐜𝐭𝐮𝐫𝐢𝐧𝐠 𝐰𝐢𝐭𝐡 ‘𝐏𝐡𝐲𝐬𝐢𝐜𝐬-𝐈𝐧𝐟𝐨𝐫𝐦𝐞𝐝 𝐀𝐈’ 🚀🚀🚀 | Via VentureBeat | June 20, 2024 | @VINCI_Digital
Los Angeles-based GrayMatter Robotics , a startup addressing some of the hardest problems in 𝐦𝐚𝐧𝐮𝐟𝐚𝐜𝐭𝐮𝐫𝐢𝐧𝐠 𝐰𝐢𝐭𝐡 𝐀𝐈-𝐩𝐨𝐰𝐞𝐫𝐞𝐝 𝐫𝐨𝐛𝐨𝐭𝐬, today announced it has raised $𝟒𝟓 𝐦𝐢𝐥𝐥𝐢𝐨𝐧 in a series B round of funding.
🔹 The investment takes the total capital raised by the company to $70 million and has been led by Wellington Management with participation from multiple new and existing investors.
🔹While robotic automation has been around for a long time, with companies like Apple using it in different functions of the assembly line, GrayMatter Robotics is pioneering what it describes as “𝐩𝐡𝐲𝐬𝐢𝐜𝐬-𝐢𝐧𝐟𝐨𝐫𝐦𝐞𝐝 𝐀𝐈” — 𝐚 𝐭𝐞𝐜𝐡𝐧𝐨𝐥𝐨𝐠𝐲 𝐭𝐡𝐚𝐭 𝐞𝐧𝐚𝐛𝐥𝐞𝐬 𝐫𝐨𝐛𝐨𝐭𝐬 𝐭𝐨 𝐬𝐞𝐥𝐟-𝐩𝐫𝐨𝐠𝐫𝐚𝐦 𝐚𝐧𝐝 𝐡𝐚𝐧𝐝𝐥𝐞 𝐡𝐢𝐠𝐡-𝐦𝐢𝐱, 𝐡𝐢𝐠𝐡-𝐯𝐚𝐫𝐢𝐚𝐛𝐢𝐥𝐢𝐭𝐲 𝐦𝐚𝐧𝐮𝐟𝐚𝐜𝐭𝐮𝐫𝐢𝐧𝐠 𝐞𝐧𝐯𝐢𝐫𝐨𝐧𝐦𝐞𝐧𝐭𝐬.
🔹This is essentially the heart of the company, which has seen significant growth since its launch in 2020.
Source: https://bit.ly/45BMl8G
----------------------------------------------
#iot #ai #genai #robotics #aipoweredrobots #physicsinformedai #manufacturing
𝐈𝐧𝐝𝐮𝐬𝐭𝐫𝐢𝐚𝐥 𝐓𝐞𝐜𝐡 𝐏𝐮𝐭𝐭𝐢𝐧𝐠 𝐭𝐡𝐞 𝐌𝐚𝐜𝐡𝐢𝐧𝐞 𝐢𝐧 𝐌𝐚𝐜𝐡𝐢𝐧𝐞 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 >>> 𝐄𝐮𝐫𝐨𝐩𝐞𝐚𝐧 𝐈𝐧𝐝𝐮𝐬𝐭𝐫𝐢𝐚𝐥 𝐓𝐞𝐜𝐡 𝐌𝐚𝐫𝐤𝐞𝐭 𝐌𝐚𝐩: 61 𝘴𝘵𝘢𝘳𝘵𝘶𝘱𝘴 𝘴𝘦𝘭𝘦𝘤𝘵𝘦𝘥 𝘣𝘺 𝘚𝘪𝘧𝘵𝘦𝘥, 𝘧𝘳𝘰𝘮 𝘱𝘳𝘦-𝘴𝘦𝘦𝘥 𝘵𝘰 𝘚𝘦𝘳𝘪𝘦𝘴 𝘈 🚀🚀🚀 | Via Sifted / Financial Times | June 6, 2024 | @VINCI_Digital
🔹Germany’s Daedalus, founded by one of OpenAI’s first engineering hires, raised $21m in February to develop its vision for AI-controlled factories.
🔹Belgium’s Robovision inked a $42m deal in March to bring computer vision intelligence to industrial machinery, while Switzerland’s EthonAI raised a $16.5m Series A in May for software that predicts maintenance needs and catches quality issues early.
🔹Further evidence of a swell in AI manufacturing can be seen in the $28.5m Series B for UK-founded Oden Technologies, now based in New York, in April, while Apple a month earlier acquired Canadian startup DarwinAI, which visually inspects components during the manufacturing process.
🔹So-called predictive maintenance startups in Europe are suddenly experiencing VC love. aiomatic Solutions — based in Germany, as so many of these manufacturing startups are — raised €2m in January for a system that detects potential machine faults early and helps reduce downtime. Tomorrow Things, also German, raised €1.5m in March for a similar product.
🔹At the same time, we should expect to see more digital assistants enter factories. Stuttgart-based Assemblio creates animated step-by-step assembly guides for workers — the company raised €2.1m in February.
🔹Startups are also mining factory data to develop new safety tools: Sweden’s Buddywise raised €3.5m in March for a system that keeps a logbook of events marked urgent in factories, like when a forklift travels into an area it shouldn’t.
Source: https://bit.ly/3VFDz6p
----------------------------------------------
#industrial #iot #iiot #ai #genai #industrialai #industrialtech
OpenAI ’𝐬 𝐍𝐞𝐰 𝐆𝐏𝐓-𝟒𝐨 𝐂𝐚𝐧 𝐁𝐞 𝐒𝐚𝐫𝐜𝐚𝐬𝐭𝐢𝐜, 𝐒𝐢𝐧𝐠 𝐇𝐚𝐩𝐩𝐲 𝐁𝐢𝐫𝐭𝐡𝐝𝐚𝐲 𝐚𝐧𝐝 𝐓𝐞𝐚𝐜𝐡 𝐌𝐚𝐭𝐡: 𝘖𝘱𝘦𝘯𝘈𝘐’𝘴 𝘯𝘦𝘸 𝘎𝘗𝘛-4𝘰 𝘪𝘴 𝘩𝘦𝘳𝘦 — 𝘢𝘯𝘥 𝘪𝘵 𝘤𝘢𝘯 𝘭𝘢𝘶𝘨𝘩 𝘢𝘵 𝘣𝘢𝘥 𝘫𝘰𝘬𝘦𝘴 (𝘢𝘯𝘥 𝘤𝘳𝘢𝘤𝘬 𝘪𝘵𝘴 𝘰𝘸𝘯), 𝘴𝘪𝘯𝘨 𝘪𝘯 𝘵𝘶𝘯𝘦 𝘢𝘯𝘥 𝘩𝘦𝘭𝘱 𝘩𝘢𝘪𝘭 𝘓𝘰𝘯𝘥𝘰𝘯 𝘤𝘢𝘣𝘴 𝘸𝘪𝘵𝘩 𝘳𝘦𝘢𝘭𝘪𝘴𝘵𝘪𝘤 𝘦𝘮𝘰𝘵𝘪𝘰𝘯 𝘢𝘯𝘥 𝘢𝘮𝘪𝘥𝘴𝘵 𝘳𝘦𝘨𝘶𝘭𝘢𝘳 𝘩𝘶𝘮𝘢𝘯 𝘪𝘯𝘵𝘦𝘳𝘳𝘶𝘱𝘵𝘪𝘰𝘯. 🚀🚀🚀 | Via VentureBeat | May 13, 2024 | @VINCI_Digital
OpenAI today [5/13] released 16 videos of GPT-4o (short for GPT-4 omni) in action, showing the multimodal foundation large language model (LLM) interacting with the world in male and female voices in real time based on audio, visual and text inputs.
🔹For instance, after correctly identifying that the human it was speaking with (OpenAI president Greg Brockman) was preparing to make a big announcement — based on his professional attire and the presence of studio lights and a microphone — the model was informed that it was the subject.
🔹A female voice replied, seemingly coyly:
“𝘛𝘩𝘦 𝘢𝘯𝘯𝘰𝘶𝘯𝘤𝘦𝘮𝘦𝘯𝘵 𝘪𝘴 𝘢𝘣𝘰𝘶𝘵 𝘮𝘦? 𝘞𝘦𝘭𝘭 𝘤𝘰𝘭𝘰𝘳 𝘮𝘦 𝘪𝘯𝘵𝘳𝘪𝘨𝘶𝘦𝘥. 𝘠𝘰𝘶’𝘷𝘦 𝘨𝘰𝘵 𝘮𝘦 𝘰𝘯 𝘵𝘩𝘦 𝘦𝘥𝘨𝘦 𝘰𝘧 𝘮𝘺…𝘸𝘦𝘭𝘭, 𝘐 𝘥𝘰𝘯’𝘵 𝘳𝘦𝘢𝘭𝘭𝘺 𝘩𝘢𝘷𝘦 𝘢 𝘴𝘦𝘢𝘵 𝘣𝘶𝘵 𝘺𝘰𝘶 𝘨𝘦𝘵 𝘵𝘩𝘦 𝘪𝘥𝘦𝘢.”
Videos: https://lnkd.in/dfx4zT2h
Source: https://lnkd.in/dJjBbgXJ
----------------------------------------------
#iot #ai #genai #chatgpt #chatgpt4o #her
Microsoft 𝐂𝐫𝐞𝐚𝐭𝐞𝐬 𝐓𝐨𝐩 𝐒𝐞𝐜𝐫𝐞𝐭 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 𝐒𝐞𝐫𝐯𝐢𝐜𝐞 𝐟𝐨𝐫 𝐔𝐒 𝐒𝐩𝐢𝐞𝐬:>>> 𝘗𝘳𝘰𝘥𝘶𝘤𝘵 𝘸𝘦𝘯𝘵 𝘭𝘪𝘷𝘦 𝘛𝘩𝘶𝘳𝘴𝘥𝘢𝘺 [5/2] 𝘢𝘧𝘵𝘦𝘳 𝘰𝘷𝘦𝘳𝘩𝘢𝘶𝘭𝘪𝘯𝘨 𝘢 𝘴𝘶𝘱𝘦𝘳𝘤𝘰𝘮𝘱𝘶𝘵𝘦𝘳>>> 𝘜𝘚 𝘴𝘱𝘺 𝘢𝘨𝘦𝘯𝘤𝘪𝘦𝘴 𝘢𝘳𝘦 𝘬𝘦𝘦𝘯 𝘵𝘰 𝘩𝘢𝘳𝘯𝘦𝘴𝘴 𝘨𝘦𝘯𝘦𝘳𝘢𝘵𝘪𝘷𝘦 𝘈𝘐 𝘵𝘦𝘤𝘩𝘯𝘰𝘭𝘰𝘨𝘺 🚀🚀🚀 | Via Bloomberg | May 7, 2024 | @VINCI_Digital
Microsoft Corp. has deployed a generative AI model entirely divorced from the internet, saying US intelligence agencies can now safely harness the powerful technology to analyze top-secret information.
🔹It’s the first time a major large language model has operated fully separated from the internet, a senior executive at the US company said. Most AI models including OpenAI’s ChatGPT rely on cloud services to learn and infer patterns from data, but Microsoft wanted to deliver a truly secure system to the US intelligence community.
🔹 Spy agencies around the world want generative AI to help them understand and analyze the growing amounts of classified information generated daily, but must balance turning to large language models with the risk that data could leak into the open — or get deliberately hacked.
Source: https://bloom.bg/3wrrDv4
----------------------------------------------
#iot #ai #genai #llms #secretservices #spyagencies #intelligence
𝐁𝐮𝐢𝐥𝐝𝐢𝐧𝐠 𝐭𝐡𝐞 𝐅𝐮𝐭𝐮𝐫𝐞: 𝐀 𝐃𝐞𝐞𝐩 𝐃𝐢𝐯𝐞 𝐈𝐧𝐭𝐨 𝐭𝐡𝐞 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 𝐀𝐩𝐩 𝐈𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 𝐒𝐭𝐚𝐜𝐤 🚀🚀🚀 | Via Sapphire Ventures | May 29, 2024 | @VINCI_Digital
Recommended by LinkedIn
In response to the [current] market conditions, an entire 𝐞𝐜𝐨𝐬𝐲𝐬𝐭𝐞𝐦 of 𝐧𝐞𝐰 𝐭𝐨𝐨𝐥𝐬 and associated 𝐬𝐭𝐚𝐫𝐭𝐮𝐩𝐬 has rapidly emerged, creating an environment that can be difficult to navigate as decision makers consider questions of:
>>> 𝐛𝐮𝐢𝐥𝐝 𝐯𝐬. 𝐛𝐮𝐲
>>> 𝐩𝐥𝐚𝐭𝐟𝐨𝐫𝐦 𝐯𝐬. 𝐛𝐞𝐬𝐭-𝐨𝐟-𝐛𝐫𝐞𝐞𝐝
>>> 𝐢𝐧𝐜𝐮𝐦𝐛𝐞𝐧𝐭 𝐯𝐬. 𝐬𝐭𝐚𝐫𝐭𝐮𝐩𝐬.
🔹This 𝐆𝐞𝐧𝐀𝐈 𝐀𝐩𝐩 𝐈𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 𝐒𝐭𝐚𝐜𝐤 𝐦𝐚𝐫𝐤𝐞𝐭 𝐦𝐚𝐩 is a guide for builders, providing a comprehensive view of the functional domains, both above and below the foundational model API line and associated startups that are starting to define this new application pattern.
🔹We have sorted companies into groupings based on initial and/or historical market focus but note that—especially given the speed of innovation in this space—providers are expanding their product portfolios rapidly and, in some instances, could be represented in multiple areas.
Source: https://lnkd.in/dbBRNSiZ
----------------------------------------------
#iot #ai #genai #app #infrastructure #stack #startups
Waabi 𝐑𝐚𝐢𝐬𝐞𝐬 $𝟐𝟎𝟎𝐌 𝐭𝐨 𝐋𝐚𝐮𝐧𝐜𝐡 𝐒𝐞𝐥𝐟-𝐃𝐫𝐢𝐯𝐢𝐧𝐠 𝐓𝐫𝐮𝐜𝐤𝐬 𝐛𝐲 𝟐𝟎𝟐𝟓: 𝘞𝘢𝘢𝘣𝘪 𝘱𝘪𝘰𝘯𝘦𝘦𝘳𝘪𝘯𝘨 𝘨𝘦𝘯𝘦𝘳𝘢𝘵𝘪𝘷𝘦 𝘈𝘐 𝘧𝘰𝘳 𝘴𝘦𝘭𝘧-𝘥𝘳𝘪𝘷𝘪𝘯𝘨 𝘵𝘳𝘶𝘤𝘬𝘴, 𝘳𝘢𝘪𝘴𝘦𝘴 $200𝘔 𝘵𝘰 𝘤𝘰𝘮𝘮𝘦𝘳𝘤𝘪𝘢𝘭𝘭𝘺 𝘭𝘢𝘶𝘯𝘤𝘩 𝘧𝘶𝘭𝘭𝘺 𝘥𝘳𝘪𝘷𝘦𝘳𝘭𝘦𝘴𝘴 𝘵𝘳𝘶𝘤𝘬𝘴 𝘰𝘯 𝘳𝘰𝘢𝘥𝘴 𝘪𝘯 2025. 🚀🚀🚀 | Via FreightCaviar | June 19, 2024 | @VINCI_Digital
Waabi , the AI company pioneering self-driving truck tech, just landed a whopping $𝟐𝟎𝟎𝐌 𝐒𝐞𝐫𝐢𝐞𝐬 𝐁 round led by Uber and Khosla Ventures . They've got ambitious plans to get autonomous semi-trucks hauling loads with nobody behind the wheel by next year.
𝐓𝐡𝐞 𝐒𝐞𝐜𝐫𝐞𝐭 𝐒𝐚𝐮𝐜𝐞 >>> 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈
🔹 Waabi has built a powerful AI that can "𝘳𝘦𝘢𝘴𝘰𝘯" like humans.
🔹 Their AI driver generalizes to any situation, not just what's in training data.
🔹 Requires way less data and compute resources than other approaches.
🔹 Paired with a hyper-realistic closed-loop simulator called "𝘞𝘢𝘢𝘣𝘪 𝘞𝘰𝘳𝘭𝘥."
Source: https://bit.ly/3Vz9nbI
----------------------------------------------
#iot #ai #genai #trucking #selfdrivingtrucks #autonomoustrucking
𝐌𝐚𝐤𝐞 𝐑𝐨𝐨𝐦 𝐟𝐨𝐫 𝐑𝐀𝐆: 𝐇𝐨𝐰 𝐆𝐞𝐧 𝐀𝐈'𝐬 𝐁𝐚𝐥𝐚𝐧𝐜𝐞 𝐨𝐟 𝐏𝐨𝐰𝐞𝐫 𝐢𝐬 𝐒𝐡𝐢𝐟𝐭𝐢𝐧𝐠: 𝘈𝘐'𝘴 𝑹𝑨𝑮 𝘪𝘴 𝘯𝘰𝘵 𝘢 𝘱𝘢𝘯𝘢𝘤𝘦𝘢 𝘧𝘰𝘳 𝘩𝘢𝘭𝘭𝘶𝘤𝘪𝘯𝘢𝘵𝘪𝘰𝘯𝘴, 𝘣𝘶𝘵 𝘪𝘵'𝘴 𝘩𝘦𝘳𝘦 𝘵𝘰 𝘴𝘵𝘢𝘺, 𝘸𝘩𝘪𝘤𝘩 𝘮𝘢𝘺 𝘮𝘦𝘢𝘯 𝘶𝘭𝘵𝘪𝘮𝘢𝘵𝘦𝘭𝘺 𝘢𝘥𝘢𝘱𝘵𝘪𝘯𝘨 𝘵𝘩𝘦 𝘵𝘳𝘢𝘪𝘯𝘪𝘯𝘨 𝘰𝘧 𝘭𝘢𝘳𝘨𝘦 𝘭𝘢𝘯𝘨𝘶𝘢𝘨𝘦 𝘮𝘰𝘥𝘦𝘭𝘴 [𝑳𝑳𝑴𝒔] 𝘵𝘰 𝘣𝘦𝘵𝘵𝘦𝘳 𝘢𝘤𝘤𝘰𝘮𝘮𝘰𝘥𝘢𝘵𝘦 𝑹𝑨𝑮. 🚀🚀🚀 | Via ZDNET / By Tiernan Ray | June 3, 2024 | @VINCI_Digital
A growing body of work suggests the technology of retrieval-augmented generation, or 𝐑𝐀𝐆, could be pivotal in shaping the battle between large language models (𝐋𝐋𝐌𝐬).
🔹𝐑𝐀𝐆 is the practice of having an 𝐋𝐋𝐌 respond to a prompt by sending a request to some external data source, such as a "𝒗𝒆𝒄𝒕𝒐𝒓 𝒅𝒂𝒕𝒂𝒃𝒂𝒔𝒆", and retrieve authoritative data. The most common use of 𝐑𝐀𝐆 is to reduce the propensity of 𝐋𝐋𝐌𝐬 to produce "𝒉𝒂𝒍𝒍𝒖𝒄𝒊𝒏𝒂𝒕𝒊𝒐𝒏𝒔", where a model asserts falsehoods confidently.
🔹Commercial software vendors, such as search software maker Elastic, and "𝒗𝒆𝒄𝒕𝒐𝒓" database vendor Pinecone, are rushing to sell programs that let companies hook up to databases and retrieve authoritative answers grounded in, for example, a company's product data.
🔹What's retrieved can take many forms, including documents from a document database, images from a picture file or video, or pieces of code from a software development code repository.
----------------------------------------------
#iot #ai #genai #llms #rag #retrievalaugmentedgeneration #vectordatabase
𝐓𝐡𝐞 𝐄𝐧𝐭𝐞𝐫𝐩𝐫𝐢𝐬𝐞 𝐀𝐈 𝐑𝐨𝐚𝐝𝐦𝐚𝐩: 𝐇𝐨𝐰 𝐭𝐡𝐞 𝐀𝐈 𝐃𝐞𝐯𝐞𝐥𝐨𝐩𝐦𝐞𝐧𝐭 𝐏𝐥𝐚𝐭𝐟𝐨𝐫𝐦 𝐋𝐚𝐧𝐝𝐬𝐜𝐚𝐩𝐞 𝐡𝐚𝐬 𝐓𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐞𝐝, 𝐂𝐡𝐚𝐧𝐠𝐢𝐧𝐠 𝐇𝐨𝐰 𝐁𝐮𝐲𝐞𝐫𝐬 𝐀𝐬𝐬𝐞𝐬𝐬 𝐑𝐎𝐈, 𝐔𝐬𝐞 𝐂𝐚𝐬𝐞𝐬, 𝐚𝐧𝐝 𝐌𝐨𝐫𝐞 | 𝘞𝘦 𝘮𝘪𝘯𝘦𝘥 CB Insights 𝘥𝘢𝘵𝘢 𝘢𝘯𝘥 𝘴𝘱𝘰𝘬𝘦 𝘸𝘪𝘵𝘩 50+ 𝘣𝘶𝘺𝘦𝘳𝘴 𝘰𝘧 𝘈𝘐 𝘥𝘦𝘷𝘦𝘭𝘰𝘱𝘮𝘦𝘯𝘵 𝘱𝘭𝘢𝘵𝘧𝘰𝘳𝘮𝘴 — 𝘪𝘯𝘤𝘭𝘶𝘥𝘪𝘯𝘨 Databricks , Hugging Face , 𝘢𝘯𝘥 Scale AI — 𝘵𝘰 𝘶𝘯𝘥𝘦𝘳𝘴𝘵𝘢𝘯𝘥 𝘩𝘰𝘸 𝘵𝘩𝘦𝘺’𝘳𝘦 𝘥𝘦𝘱𝘭𝘰𝘺𝘪𝘯𝘨 𝘈𝘐 𝘮𝘰𝘥𝘦𝘭𝘴, 𝘸𝘩𝘢𝘵 𝘵𝘩𝘦𝘺’𝘳𝘦 𝘱𝘢𝘺𝘪𝘯𝘨, 𝘢𝘯𝘥 𝘸𝘩𝘢𝘵 𝘵𝘩𝘦 𝘧𝘶𝘵𝘶𝘳𝘦 𝘩𝘰𝘭𝘥𝘴 𝘧𝘰𝘳 𝘵𝘩𝘦 𝘭𝘢𝘯𝘥𝘴𝘤𝘢𝘱𝘦. 🚀🚀🚀 | Via CB Insights | June 21, 2024 | @VINCI_Digital
𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 𝐡𝐚𝐬 𝐦𝐚𝐫𝐤𝐞𝐝 𝐚 𝐧𝐞𝐰 𝐞𝐫𝐚 𝐢𝐧 𝐀𝐈 𝐝𝐞𝐯𝐞𝐥𝐨𝐩𝐦𝐞𝐧𝐭
🔹Increasingly sophisticated models are opening up new applications that enterprises are eager to harness.
🔹AI development platforms — which enable enterprises to manage the AI lifecycle, from data preparation to model deployment and continuous monitoring — are responding to the call.
🔹These one-stop shop platforms, from enterprise machine learning (ML) players (Dataiku, H2O.ai) to big tech products (Google Cloud Vertex AI, Amazon SageMaker and Bedrock), are jumping on the opportunity to adapt their offerings for the genAI era and capture more enterprise AI spend.
Source: https://lnkd.in/dF85h2vE
----------------------------------------------
#iot #ai #genai #enterprise #aidevelopment #platforms #roi
Meta ’𝐬 𝐍𝐞𝐰 𝐌𝐮𝐥𝐭𝐢-𝐓𝐨𝐤𝐞𝐧 𝐏𝐫𝐞𝐝𝐢𝐜𝐭𝐢𝐨𝐧 𝐌𝐚𝐤𝐞𝐬 𝐀𝐈 𝐌𝐨𝐝𝐞𝐥𝐬 𝐔𝐩 𝐭𝐨 𝟑𝐗 𝐅𝐚𝐬𝐭𝐞𝐫: 𝘐𝘯 𝘢 𝘳𝘦𝘤𝘦𝘯𝘵 𝘴𝘵𝘶𝘥𝘺, 𝘳𝘦𝘴𝘦𝘢𝘳𝘤𝘩𝘦𝘳𝘴 𝘢𝘵 Meta , École nationale des ponts et chaussées 𝘢𝘯𝘥 Université Paris-Saclay 𝘴𝘶𝘨𝘨𝘦𝘴𝘵 𝘪𝘮𝘱𝘳𝘰𝘷𝘪𝘯𝘨 𝘵𝘩𝘦 𝘢𝘤𝘤𝘶𝘳𝘢𝘤𝘺 𝘢𝘯𝘥 𝘴𝘱𝘦𝘦𝘥 𝘰𝘧 𝘈𝘐 𝘭𝘢𝘳𝘨𝘦 𝘭𝘢𝘯𝘨𝘶𝘢𝘨𝘦 𝘮𝘰𝘥𝘦𝘭𝘴 (𝘓𝘓𝘔𝘴) 𝘣𝘺 𝘮𝘢𝘬𝘪𝘯𝘨 𝘵𝘩𝘦𝘮 𝘱𝘳𝘦𝘥𝘪𝘤𝘵 𝘮𝘶𝘭𝘵𝘪𝘱𝘭𝘦 𝘵𝘰𝘬𝘦𝘯𝘴 𝘴𝘪𝘮𝘶𝘭𝘵𝘢𝘯𝘦𝘰𝘶𝘴𝘭𝘺. 🚀🚀🚀 | Via VentureBeat | May 6, 2024 | @VINCI_Digital
The new study by Meta is predicated on the hypothesis that “𝘵𝘳𝘢𝘪𝘯𝘪𝘯𝘨 𝘭𝘢𝘯𝘨𝘶𝘢𝘨𝘦 𝘮𝘰𝘥𝘦𝘭𝘴 𝘵𝘰 𝘱𝘳𝘦𝘥𝘪𝘤𝘵 𝘮𝘶𝘭𝘵𝘪𝘱𝘭𝘦 𝘧𝘶𝘵𝘶𝘳𝘦 𝘵𝘰𝘬𝘦𝘯𝘴 𝘢𝘵 𝘰𝘯𝘤𝘦 𝘳𝘦𝘴𝘶𝘭𝘵𝘴 𝘪𝘯 𝘩𝘪𝘨𝘩𝘦𝘳 𝘴𝘢𝘮𝘱𝘭𝘦 𝘦𝘧𝘧𝘪𝘤𝘪𝘦𝘯𝘤𝘺.”
🔹𝐌𝐮𝐥𝐭𝐢-𝐭𝐨𝐤𝐞𝐧 𝐩𝐫𝐞𝐝𝐢𝐜𝐭𝐢𝐨𝐧 instructs the LLM to predict several future tokens from each position in the training corpora at the same time. The researchers propose a simple multi-token prediction architecture that does not require extra training time or memory overhead.
🔹 The 𝐦𝐮𝐥𝐭𝐢-𝐭𝐨𝐤𝐞𝐧 𝐩𝐫𝐞𝐝𝐢𝐜𝐭𝐢𝐨𝐧 language model is based on the 𝐓𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐞𝐫 architecture used in most LLMs, although with some modifications. The model uses the main structure of the 𝐓𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐞𝐫 through instead of a single output, it has multiple independent output heads, one for each token it wants to predict.
“𝘞𝘩𝘪𝘭𝘦 𝘤𝘰𝘴𝘵-𝘧𝘳𝘦𝘦 𝘢𝘯𝘥 𝘴𝘪𝘮𝘱𝘭𝘦, 𝘮𝘶𝘭𝘵𝘪-𝘵𝘰𝘬𝘦𝘯 𝘱𝘳𝘦𝘥𝘪𝘤𝘵𝘪𝘰𝘯 𝘪𝘴 𝘢𝘯 𝘦𝘧𝘧𝘦𝘤𝘵𝘪𝘷𝘦 𝘮𝘰𝘥𝘪𝘧𝘪𝘤𝘢𝘵𝘪𝘰𝘯 𝘵𝘰 𝘵𝘳𝘢𝘪𝘯 𝘴𝘵𝘳𝘰𝘯𝘨𝘦𝘳 𝘢𝘯𝘥 𝘧𝘢𝘴𝘵𝘦𝘳 𝘵𝘳𝘢𝘯𝘴𝘧𝘰𝘳𝘮𝘦𝘳 𝘮𝘰𝘥𝘦𝘭𝘴,” the researchers write.
Source: https://bit.ly/4boD5qf
----------------------------------------------
#iot #ai #genai #multitoken #prediction #llms #transformer
𝐇𝐢, 𝐀𝐈: 𝐎𝐮𝐫 𝐓𝐡𝐞𝐬𝐢𝐬 𝐨𝐧 𝐀𝐈 𝐕𝐨𝐢𝐜𝐞 𝐀𝐠𝐞𝐧𝐭𝐬 >>> 𝘕𝘰𝘸 𝘪𝘴 𝘵𝘩𝘦 𝘵𝘪𝘮𝘦 𝘵𝘰 𝘳𝘦𝘪𝘯𝘷𝘦𝘯𝘵 𝘵𝘩𝘦 𝘱𝘩𝘰𝘯𝘦 𝘤𝘢𝘭𝘭. 𝘛𝘩𝘢𝘯𝘬𝘴 𝘵𝘰 𝘨𝘦𝘯 𝘈𝘐, 𝘯𝘰 𝘩𝘶𝘮𝘢𝘯 𝘸𝘪𝘭𝘭 𝘦𝘷𝘦𝘳 𝘢𝘨𝘢𝘪𝘯 𝘩𝘢𝘷𝘦 𝘵𝘰 𝘮𝘢𝘬𝘦 𝘢 𝘤𝘢𝘭𝘭. 𝘏𝘶𝘮𝘢𝘯𝘴 𝘸𝘪𝘭𝘭 𝘴𝘱𝘦𝘯𝘥 𝘵𝘪𝘮𝘦 𝘰𝘯 𝘵𝘩𝘦 𝘱𝘩𝘰𝘯𝘦 𝘰𝘯𝘭𝘺 𝘸𝘩𝘦𝘯 𝘢 𝘤𝘢𝘭𝘭 𝘩𝘢𝘴 𝘷𝘢𝘭𝘶𝘦 𝘵𝘰 𝘵𝘩𝘦𝘮. 🚀🚀🚀 | Via Andreessen Horowitz / By Olivia Moore & Anish Acharya | May 29, 2024 | @VINCI_Digital
𝐁𝟐𝐁 𝐀𝐠𝐞𝐧𝐭𝐬: 𝐎𝐮𝐫 𝐓𝐡𝐞𝐬𝐢𝐬
We are transitioning from 1.0 AI voice (phone tree) → 2.0 wave of AI voice (LLM-based). 2.0 companies have been emerging in the last 6 months or so. 1.0 companies may be more accurate now, but the 2.0 approach should be much more scalable and accurate in the long term.
🔹There is unlikely to be one horizontal model or platform that works across all types of enterprise voice agents. There are a few key differences across verticals:
(1) call types, tones, and structures
(2) integrations and processes
(3) GTM and “𝘬𝘪𝘭𝘭𝘦𝘳 𝘧𝘦𝘢𝘵𝘶𝘳𝘦𝘴.”
🔹This is likely to mean an explosion of vertical agents that are highly opinionated in the UI. This requires founding teams with deep domain expertise or interest. Labor is the #1 cost center for many businesses — TAM is large for companies that “𝘨𝘦𝘵 𝘪𝘵 𝘳𝘪𝘨𝘩𝘵.”
The most near-term opportunities may be in industries that live and die by phone appointments, have significant labor shortages, and have low call complexity. As agents become more sophisticated, they will be able to tackle more complex calls.
Source: https://bit.ly/4e8Jfgg
----------------------------------------------
#iot #ai #genai #llms #aiagents #voiceagents
Bio: 🔹Fabio Bottacci is a relationship builder, creative problem solver, and strategic thinker. Senior industrial executive and former strategic consultant, he acquired a solid background in large multinational organizations across Brazil, US, and Western Europe. He is known for his ability to deliver results despite ambiguity and obstacles, to build bridges between people, and to manage conflict and negotiations.
Fabio graduated in Economics & Business Administration from Università Bocconi and from Stanford University School of Engineering in Internet of Things [IoT]. Recently, Fabio has successfully completed a Private Equity & Venture Capital course by SDA Bocconi [Final Grade: 94%], and two accredited certifications: Generative AI by Databricks Academy and Generative AI by Andrew Ng, founder DeepLearning.AI [Final Grade: 96%].
He began his career at Accenture Italia/Accenture, strategy practice, while attending MBA courses. He then moved to Brazil, where he consistently proved, during more than 20 years of professional experience, strong clients' network, industry knowledge and business development expertise in the automotive, oil and gas, energy, and utilities verticals.
Since 2015, he has been the founder & CEO @VINCI Digital | IIoT + GenAI Strategic Advisory 🚀 , being recognized internationally as a thought leader by well-known organizations, such as the World Economic Forum, the IOT Solutions World Congress, and the BNDES (Brazilian Federal Development Bank).
Recently, Fabio has been contracted by Bain & Company as Senior Advisor to its Advanced Manufacturing & Services global practice, with a particular focus on Industrial IoT [IIoT] + AI / Generative AI's end-to-end solutions and their actual deployment, at-scale, within the Pulp & Paper, Oil & Energy / Utilities, Automotive, and Industrial Equipment verticals.
In 2023, he had joined the European Research Executive Agency (REA) as an Industrial IoT and Generative AI Expert / Evaluator for the HORIZON Europe Programme of the European Commission, and Kobo Funds as Venture Partner, focused on Deeptech Seed/Series A startups, helping investors work successfully with entrepreneurs.
Fabio's mission is to help startups, SME, and corporations to thrive within the current digital transformation environment, by increasing productivity, developing new business models, and delivering actual results / ROI in months, not years!
🔹Liked this newsletter? 🔹Want to see more? 🔹Subscribe!