Tim Stahl

AI as a Game Changer: The New Driving Force of the Automotive Industry Authors:   - Dr. Nils Schaupensteiner (Associated Partner, MHP GmbH)   - Augustin Friedel 🚲 🛴 🛵 🚘 (Senior Manager, MHP GmbH)   - Matthias Borch (Consultant: Data Science & AI, MHP GmbH)   - Marcus Willand (Partner, MHP GmbH)   - Patrick Ruhland (Consultant Mobility Transformation , MHP GmbH)  This study, published by MHP – A Porsche Company, examines the transformative role of #ArtificialIntelligence (#AI) in reshaping the #automotiveindustry. The authors provide an in-depth analysis of AI's #applications, its disruptive potential across the #automotivevaluechain, and the strategic imperatives for #stakeholders. Key Insights: 1. AI’s Growth in the Automotive Sector:   - The market size of AI-based systems in the #automotiveindustry is projected to grow from $3.9 billion in 2024 to $35.7 billion by 2033, reflecting an annual #growth rate of 28%. This surge underscores AI's role as a pivotal platform shift following the adoption of cloud and #SaaS technologies. 2. Applications Across the Value Chain:   - AI is driving innovation in areas such as:    - Advanced Driver Assistance Systems (#ADAS) and autonomous driving, with companies like Tesla and XPeng leading developments in end-to-end #neuralnetworkapplications.    - Manufacturing and Quality Control, leveraging AI for predictive maintenance and process optimization, achieving efficiency gains of up to 25% in leading factories like BMW's iFACTORY.    - Customer Experience, with #personalizedvirtualassistants and in-vehicle AI enabling features like intelligent route planning and predictive diagnostics. 3. Investment Trends:   - Significant AI investment is concentrated in foundational models and applications. In 2024, $14 billion was invested in AI model development, constituting 62% of AI-related investments. However, the automotive industry’s current #investment in AI teams and budgets is below average, with only 30% of companies reporting dedicated resources. 4. Strategic Recommendations:   - Prioritize digital transformation as a prerequisite for AI implementation.   - Invest in AI infrastructure and data-sharing frameworks to enhance innovation while reducing development costs.   - Foster collaborations with technology firms and hyperscalers to accelerate progress in AI applications, particularly in software-defined vehicles. Conclusion: AI is poised to revolutionize the automotive industry, from enabling fully #autonomous vehicles to optimizing supply chains and enhancing customer experiences. However, realizing its full potential requires strategic foresight, investment in digitalization, and effective collaboration across the ecosystem. The report emphasizes that companies adapting early to this paradigm shift will secure competitive advantages in the evolving automotive landscape.

10

3 Kommentare

🚗 𝐑𝐨𝐛𝐨𝐭𝐚𝐱𝐢𝐬- 𝐓𝐡𝐞 𝐟𝐮𝐭𝐮𝐫𝐞 𝐨𝐟 𝐫𝐢𝐝𝐞 𝐡𝐚𝐢𝐥𝐢𝐧𝐠? Imagine a world where driving is a relic of the past- a skill preserved only in history books. Waymo, once a secret “self-driving car project” at Google, has made it a reality, such that #GenAlpha (Born 2010-2025) may never need a driver’s license in the USA at least. What began in 2009 with a bold ambition to transform transportation has evolved into a revolution on wheels. Waymo made headlines with the first fully-autonomous public ride in 2015 and broke new ground in 2018 with Waymo One- the world’s first commercial autonomous ride-hailing service. 🌟 The End of Human Driving As of 2024, Waymo, valued at $45B, is the monopoly player dominating the driverless ride-hailing market in the USA, with 150K paid trips per week in Phoenix, SF, and LA. Come 2025, Waymo is well-positioned to solidify its lead starting with Austin, Miami and Atlanta next. It is a race against time to stay ahead of China in regulatory approvals. The use case is compelling: driverless ride-hailing costs just $0.70 per mile, compared to conventional Uber rides at $2.40 per mile. Given that 40K+ people lose  their lives in motor accidents in the USA alone, this revolution will make our roads much safer besides the ease and the convenience.   That said, Tesla’s computer vision technology is set to challenge Waymo, but not for at least 2 years, according to Tesla itself. And, as we know, Tesla rarely delivers on time. 🚨 Impact on Uber, & traditional Ride-Hailing Waymo's development presents major challenges to Uber. Yet Uber has significant defensibility. With 161 Million monthly active platform customers clocking 3 Billion trips in the last quarter, across 70 countries, Uber is the gorilla of ride hailing. More importantly, outside of the USA and China, there are hardly any robo-taxi initiatives worth writing home about. Since Uber is not present in China, USA remains its only critical market for transitioning to robo-taxis. To this end, it has entered into a landmark agreement with Waymo. 🤝The Uber-Waymo Partnership With Waymo’s integration with Uber, riders can hail Waymo cars directly from the Uber app, giving Waymo a head start over future competition from Tesla and Uber with clear access to Waymo robo-taxi. Yet all these are not given. Last week, Waymo announced  expansion into Miami through Moove, bypassing Uber as a partner. Ironically Uber led Moove's $100M Series B, giving them a proxy play. At DECACORN, we believe personal experience is the best way to carry out due diligence. We tried a Waymo in SF streets this fall- what an experience! With Ride hailing expected to grow 63.5% CAGR between 2024-32, this is a generational transformation to safer streets and cheaper rides making ride hailing a superior experience full of ease and convenience! If you're in the coverage area, book a ride and see for yourself. Let me know what you think!

42

4 Kommentare

I'm excited to announce that I'll be speaking at DLR Conference 2024 about a paper I co-authored with my former student Ricardo Blasaditsch titled "Enhancing Air-to-Air Combat Effectiveness Through Optimised Missile Launch Timing". In our study, we present a novel approach that combines supervised machine learning with advanced mathematical modelling to optimise missile engagement strategies in air-to-air combat. By analysing data from Airbus' Air Combat Evaluation (ACE) simulation, we evaluated several models and found that XGBoost provided the best balance of accuracy and speed for real-time decision making. Our research highlights the potential of AI-driven solutions to improve tactical decision-making in future air combat scenarios. If you haven't registered yet, you can do so here: 🔗 https://dlrk2024.dglr.de Looking forward to the conference and seeing old and new faces. #DLRK2024 #AirCombat #MachineLearning #AI #Aviation #Aerospace #DefenseTechnology #TacticalDecisionMaking #Modelling #Simulation #Airbus #AirbusDefenceAndSpace #Research #Innovation

45

1 Kommentar

Good one, Wolfram Müller! https://lnkd.in/eyw_CK6e To put it into a nutshell: If you do, what everyone else is doing, you get, what everyone else gets. So the question is: Are you ok with what everyone else gets? If not, ACT NOW. Would waiting longer help you? How? NOW is the time to ACT. Take part in the dolphin study on flow and discover how to thrive in any market condition. TOC-companies do for 40+ years now, worldwide. Join the #FLOW success club: https://t1p.de/8ht7i You want to talk directly? Pick your date and time here and now: https://lnkd.in/e63VMViE To say it with Dr. Goldratt: "It's not Luck!" - CU, Wolfgang

2

Data isn't just about the flashiest new tool or the latest algorithm. It’s not just about what data can do; it’s about understanding and leveraging its value through a thriving data culture. Follow me for more insights into the agile business intelligence journey. #BusinesIntelligence #Agile #BI

36

EthonAI has raised CHF 15 million ($16.5 million) in a Series A round of funding led by Index Ventures, with participation from General Catalyst, Earlybird Venture Capital and Founderful. EthonAI can train AI models for specific use cases, for instance in electronics manufacturing where the customer supplies imagery of defect-free products and EthonAI’s Inspector software can then identify surface defects in the products during the manufacturing and assembly process. More broadly, though, EthonAI can combine data from across a company’s manufacturing setup, from sensors to line stops, and build a picture of where things are and aren’t working well — and even compare performance across multiple facilities to see where there might be room for improvement. https://lnkd.in/emphuHcC

2

So here is an evergreen again, just for a relaxing re-start after vacation (well, actually F1 is in its shutdown time 😉 ). I have already said many things about this model, but it might be worth to repeat them, at least some: - the model is from 2001/2002 and at that time it did not run on any machine, trying with NASTRAN and ABAQUS - ABAQUS was the one going further in the solution, but it could not converge witht the vertical and lateral loads, very likely due to element distortion in the contact area - The purpose of such model was to find the actual load path from ground into the RIM, that we wanted to optimise in terms of stiffness/weight, clearly with an eye on the stress level - We had to give up and find a different way to optimize the rim - When, back in 2021, I was writing the book "Computational Structural Engineering", looking into my old files I have found this model; I just took it exactly as it was and simply tried to re-run it: not only it converged ( 😯 ), but it did it in few minutes, whereas 20 years before it took hours just to come to a give up! The mentioned book can be purchased on all the Amazon marketplaces and by some bookstores that ship all around the world: Amazon: UK: https://lnkd.in/efZNZDyZ USA: https://lnkd.in/eeqtf_6g CAN: https://lnkd.in/ev-GjrP2 AUS: https://lnkd.in/gsCKYnGh FR: https://lnkd.in/e3FGfWTf ES: https://lnkd.in/exzAdMj4 NL: https://lnkd.in/euHe79c3 DE: https://lnkd.in/eMGeF2Av IT: https://lnkd.in/eAJ27yAQ PL: https://lnkd.in/exRB45dK BR: https://lnkd.in/giQPrqrw IN: https://lnkd.in/erqDzF5q MX: https://lnkd.in/gUVzZDsn Book stores: https://lnkd.in/e7J3xWp4 https://lnkd.in/gMJRpwZK https://lnkd.in/eZ6DSyyT https://lnkd.in/g3JnxDay https://lnkd.in/gDHQYHbG https://lnkd.in/gKtviqTY #ingegneriastrutturalecomputazionale #FEM #FEA #CAE #simulation #stressanalysis #structuraldesign #structuralengineering #structuralanalysis #ABAQUS #simulia #dassaultsystèmes #mechanicalengineering #mechanicaldesign #ansys #nastran #aerospaceengineering #finiteelementanalysis #altair #learningfea #analysis #finiteelement #f1 #formula1 #engineering #education #stress #aerospace #structuralengineers #motorsport #composites

268

3 Kommentare

Hello everyone! The Swiss Conference on Data Science | SDS is exciting and interesting. Now I have a question: To all colleagues interested in #robotics and #reinforcement_learning, which conferences do you attend on these topics? I look forward to your recommendations!

9

5 business days, 5 appearances: I just finished my first-ever "Conference Sprint" 🏃‍♂️. I need some advice before I talk again at this frequency... 😅 - Last Wednesday it was "The AI-Powered Developer" (in German) at #CLC_Conf in Mannheim with Daniel Hartung. - Then off to Cologne the next day for the same talk in English at #CTWebdev. - On Friday I held an online workshop "Consulting with Impact" for the internal #codecentriccrew (thanks to positive feedback, it may be developed into a public talk). - On Monday Janine Felten and I held the full-day online workshop "Professional GenAI Software Development in Practice" (in German) with 24 participants - many more than we planned. I'm glad to say that their great work and knowledge sharing in groups made it possible. - On Tuesday we held our complimentary talk "GenAI Crafting: Best Practices für professionelle GenAI-basierte Softwareentwicklung" at the same conference - #InfoDays #GenAI für Entwickler. In-between, I handled project needs which came up. Now I finally have a moment to rest and catch up... and I ended up giving input for someone's upcoming keynote 🙂. I feel quite mentally drained after a normal conference appearance. What helped me was to do significant exercise every day. I know a few speakers who do a lot of conferences - how do you identify the limit, and how do you stay fit and healthy? #PublicSpeaking

75

9 Kommentare

This week I'll do a Mentor Madness at Deep Tech Momentum by Techstars Deep Tech Berlin! DTM features the 100 coolest deep tech startups from Europe where they can learn from the best in their respective industries on workshops, keynotes and the Mentor Madness. Companies are sorted into the following deep tech tracks: ✅ Bio Builder Track ⚗ ✅ New Space and Dual Use Track 🚀 (this is where I'll be, of course) ✅ Novel AI & General Track🖥 🧠 The Mentor Madness is an intense 2-hour long session where mentors try to give as much value as possible to each startup in 10 minutes. Can't wait to participate and meet the best space startups in Europe! #techstars #mentoring #newspace #deeptech

63

6 Kommentare

🤖 Long before LLMs became a thing, DeepL.com was our go-to translator. As a German AI startup, it flew under the radar despite its value—likely because it was around before the big AI hype hit the market. Now, DeepL has launched a new language model for translation, clearly responding to competition from the likes of ChatGPT & Co. Notably, DeepL outperforms the major language models in benchmarks, showing significant improvements over its previous model. A big plus: DeepL trains its model on proprietary data - a potential advantage if you want to keep control over your data. And just so you know, DeepL doesn’t use any Pro customer data for training. In the business context, DeepL still holds an edge. It’ll be interesting to see how they evolve their business model as LLMs continue to develop. Any ideas?

27

3 Kommentare

𝗛𝗮𝘃𝗲 𝘆𝗼𝘂 𝗻𝗼𝘁𝗶𝗰𝗲𝗱 𝗵𝗼𝘄 𝘀𝗼𝗺𝗲 𝗽𝗲𝗼𝗽𝗹𝗲 𝘁𝗮𝗹𝗸 𝗮𝗯𝗼𝘂𝘁 𝘁𝗲𝘅𝘁 𝗻𝗼𝗿𝗺𝗮𝗹𝗶𝘀𝗮𝘁𝗶𝗼𝗻 𝘄𝗵𝗲𝗻 𝗶𝘁 𝗰𝗼𝗺𝗲𝘀 𝘁𝗼 𝗡𝗟𝗣 / 𝗟𝗟𝗠 𝗲𝘃𝗮𝗹𝘀? What's that? ↳ Text normalisation is converting text into a standard format with reduced variability It is not one operations, but a collection of small transformations you could choose to apply of not to your text. 𝗘𝘅𝗮𝗺𝗽𝗹𝗲𝘀 👇️ > Lowercasing text: "Hello" → "hello" > Removing punctuation "Hello, world!" → "Hello world" > Removing stopwords: Eliminate common words like "a," "an," and "the" that don't add much meaning in some contexts. > Remove extra spaces and normalise them to a single space: " Hello world " → "Hello world" > Reduce words to their base or root forms "running" → "run" > Convert numbers to words (e.g., "1" to "one") and expand abbreviations (e.g., "Dr." to "Doctor") etc... We normalise to avoid penalisation due to irrelevant variations > normalise your reference text > normalise your prediction Now compare for evaluation But !⚡ Text normalisation strategies can vary based on the problem you're solving.  Sometimes, you might want to keep punctuation, or acronyms or something else!

17

4 Kommentare

📈 The latest startupdetector report couldn't have been better for me. - the number of newly founded German startups increased in Q1 2024 - a trend reversal we waited for some time new! - Leipzig and Dresden were the cities with the biggest increase of new startups (+41%, +36%), while Jena had the most financing rounds per capita - while many regions had a smaller number of new startups, Saxony increased most by 35% - the energy sector saw the strongest increase from all industries - our Smart Infrastructure Ventures ('SIVentures') portfolio company traide AI GmbH with their founder Leonie Althaus got featured next to SpinLab - The HHL Accelerator alumnus Dr. Jannik Lockl from inContAlert - our partners from TGFS Technologiegründerfonds Sachsen are ranked #5 in the most active VCs list. Smart Infrastructure Ventures ('SIVentures') among the top investors in AI All in all a great job Arnas Bräutigam & team

61

3 Kommentare

In today's episode, extremely clear-spoken Dr. Florian Neukart — a world-leading expert on quantum computing — details real-world problems we can tackle with Quantum ML and how you can get started with QML today! Florian:  • Is Chief Product Officer and Member of the Board at Terra Quantum AG, a leading quantum-computing startup headquartered in Switzerland and Germany.  • Assistant Professor of quantum computing at Leiden University in the Netherlands.  • Holds a PhD in Quantum Computing and Machine Learning. Today’s episode gets a bit technical at some parts — particularly near the beginning — with respect to the mechanics of quantum computing and Quantum ML (QML) but, by and large, the episode should be fascinating to any interested listener. In today’s episode, Florian details:  • How a new generation of hybrid quantum-classical systems has made quantum computing practical for real-world applications.  • An overview of the available quantum computing chips, including Willow, Google’s new quantum chip, which has been making a big splash.  • The race to develop "quantum-proof" encryption.  • The breadth of real-world problems that can be tackled with quantum machine learning today. • How quantum ML could unlock personalized medicine, nuclear fusion energy and revolutionary space technologies.  • How you can get started with quantum ML yourself today. Thanks to Collibra for supporting today's episode of the "Super Data Science Podcast with Jon Krohn", enabling the show to be freely available on all major podcasting platforms as well as the video version we publish on YouTube. This is Episode #851! #superdatascience #quantumcomputing #quantumML #quantumAI #quantummachinelearning

46

7 Kommentare

🚨 The German government just denied funding to Lilium. Meanwhile, countries around the world are doubling down on innovative tech investments. Yet here we are, pulling back, citing “high risks” and budget concerns. What kind of message does this send to our startups and innovators? If we aren’t willing to back ambitious projects that could redefine transportation, who will? Germany needs a reality check. Innovation is happening NOW, and every day we hesitate, we risk falling further behind on the global stage. We need to commit to our tech talent and support the future of our economy. It’s time for a serious shift in our political mindset. Our leaders need to start betting on bold ideas and backing the next generation of German innovators—or risk watching them move elsewhere. Honestly let’s get loud about this! Engage or like if you believe in a stronger, more innovative Germany.

22

10 Kommentare

Interesting and important! Comprehensive situation awareness support... various systems are offering comparable functionalities, integrating the same layers of information. Going forward it will be important to ensure interoperability, so that first responder units using different systems can seamlessly collaborate when they join forces in case of larger incidents.

4

𝐇𝐚𝐧𝐝𝐞𝐥𝐬𝐛𝐥𝐚𝐭𝐭: “𝐀𝐈 𝐩𝐢𝐨𝐧𝐞𝐞𝐫 𝐫𝐞𝐦𝐛𝐞𝐫𝐠 𝐛𝐮𝐢𝐥𝐝𝐬 𝐨𝐧𝐞 𝐨𝐟 𝐭𝐡𝐞 𝐰𝐨𝐫𝐥𝐝’𝐬 𝐟𝐢𝐫𝐬𝐭 𝐜𝐨𝐩𝐢𝐥𝐨𝐭𝐬 𝐟𝐨𝐫 𝐢𝐧𝐝𝐮𝐬𝐭𝐫𝐢𝐚𝐥 𝐜𝐨𝐦𝐩𝐚𝐧𝐢𝐞𝐬” Over the last 12 months we have been transforming remberg into an AI-first company. Why? Because we understood that AI is the key to beating the status quo in the industrial world by 10 instead of 2-3 times. Today, a full page in Handelsblatt covers remberg’s launch of one of the world’s first AI copilots for industrial companies who service, operate or maintain assets & equipment, such as machines, energy utilities, facilities, vehicles & many more. Our AI-powered solution helps them to maximize the uptime of their assets and solve their most critical challenges: - 𝐒𝐡𝐨𝐫𝐭𝐚𝐠𝐞 𝐨𝐟 𝐬𝐤𝐢𝐥𝐥𝐞𝐝 𝐥𝐚𝐛𝐨𝐫 - 𝐈𝐧𝐜𝐫𝐞𝐚𝐬𝐞 𝐢𝐧 𝐚𝐬𝐬𝐞𝐭 & 𝐞𝐪𝐮𝐢𝐩𝐦𝐞𝐧𝐭 𝐜𝐨𝐦𝐩𝐥𝐞𝐱𝐢𝐭𝐲 - 𝐑𝐢𝐬𝐢𝐧𝐠 𝐠𝐥𝐨𝐛𝐚𝐥 𝐜𝐨𝐦𝐩𝐞𝐭𝐢𝐭𝐢𝐯𝐞 𝐩𝐫𝐞𝐬𝐬𝐮𝐫𝐞 Since Q1 this year, the adoption of digital technologies in the industrial sector accelerated significantly, which our own numbers also confirm. Our customers such as 𝐊Ä𝐑𝐂𝐇𝐄𝐑, 𝐚𝐦𝐬 𝐎𝐒𝐑𝐀𝐌, 𝐋𝐢𝐪𝐮𝐢 𝐌𝐨𝐥𝐲, 𝐀𝐜𝐭𝐞𝐦𝐢𝐮𝐦, 𝐄𝐝𝐝𝐢𝐧𝐠 and many more are already managing over 1 million different kinds of assets in our solution. We will keep on going the extra mile to provide our customers with a competitive advantage in the era of AI. In case you are operating such assets yourself, and want to see the power of AI in action, please do reach out. If you want to join the AI revolution in the industrial world, we are looking to hire bright minds across the board for remberg’s next growth phase, offering roles in: - 𝐓𝐚𝐥𝐞𝐧𝐭 𝐀𝐜𝐪𝐮𝐢𝐬𝐢𝐭𝐢𝐨𝐧 - 𝐏𝐫𝐨𝐝𝐮𝐜𝐭, 𝐃𝐞𝐬𝐢𝐠𝐧 & 𝐄𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐢𝐧𝐠 - 𝐒𝐚𝐥𝐞𝐬 - 𝐌𝐚𝐫𝐤𝐞𝐭𝐢𝐧𝐠 - 𝐃𝐞𝐬𝐢𝐠𝐧 - 𝐂𝐮𝐬𝐭𝐨𝐦𝐞𝐫 𝐒𝐮𝐜𝐜𝐞𝐬𝐬 See links in the comments: - Read the Handelsblatt Article - Watch a ~5 min video of the remberg AI copilot in action - Check out our career page #remberg #AI #copilot #industrialtech #handelsblatt

30

1 Kommentar