Nordic Data & AI Weekly: All the news from OpenAI's developer conference, the latest on AI safety & regulation, recent Nordic investment rounds, and all the latest news & resources
Recent funding, news, and resources for builders
Quick reads on value creation
Things to know, whether you are building 🛠 or investing 💰
📊 Businesses are reporting an average 3.5x return on their AI investments, according to a survey of 2,100 global business leaders and decision makers by research firm IDC.
📖 CBInsights released its “Generative AI Bible” - analysing how the tech has developed, where it's going, and which trends and players you need to watch.
⚒️ CBInsights (again) with a market map of open-source tools helping companies to build and deploy AI projects.
📃 A list of tools, research papers, and other resources to help with evaluating LLMs.
🏆 The Stanford Center for Research on Foundation Models has released the first version of its Foundation Model Transparency Index - providing a comprehensive assessment of the transparency of foundation model developers.
Recent AI funding in the Nordics & Baltics
🇸🇪 🇳🇴 🇩🇰 🇫🇮 🇪🇪 🇱🇻 🇱🇹 🇮🇸
🇸🇪 Indivd: Using surveillance cameras to provide retailers with advanced customer analytics. Raised $900k from angels.
🇳🇴 Vespa: Enabling enterprises to apply AI to their data, online, at any scale, with unbeatable performance. Raised $31m from Blossom Capital, Yahoo (link).
🇫🇮 Kuva Space: Using space-borne hyperspectral imaging and AI powered services in order to identify environmental changes and insights for increased crop yield, or creating opportunities for improved food security, defence, safety, and carbon sequestration credibility. Raised $17.6m from Voima Ventures, Springvest Oy, Nordic FoodTech VC, and Earth Venture Capital (link).
🇫🇮 Algorithmiq: Developing advanced quantum algorithms to solve complex problems in life sciences. Raised $4.25m from Wellcome Leap.
🇫🇮 Wudpecker: The new default for storing meeting knowledge - get summaries, action items, and insights from your Zoom, Google Meet, Microsoft Teams meetings. Raised $352.77k from Accelerace, Trind VC, and Sofokus (link).
🇱🇹 Cast AI: AI-driven cloud optimisation platform that slashes cloud cost, optimises DevOps, and automates disaster prevention. Raised $35m from Creandum, Vintage Investment Partners, and Uncorrelated Ventures (link).
If there is some relevant Nordic AI news or funding that is missing from this newsletter, feel free to share by replying directly or emailing nordicaiweekly@substack.com and I’ll look to spread the word next week. Thanks!
News from the wider world 🌍 🌎 🌏
📣 OpenAI had its first developer conference and made some pretty big announcements - you can see the full keynote address here and the official blog post here. A whole lot of companies representing simple GPT-wrappers will find themselves more or less eliminated, and even deeper tech players and large incumbents will be threatened (example overview of companies affected here). Announcements included:
The launch of GPTS —> custom chatbots within ChatGPT that you can create without knowing how to code, enabling users to craft personalised AI for specific uses like learning, work, or leisure, and to share these with others. More here.
The upcoming GPT store —> anyone will be able to share the GPTs they’ve created and make money through the GPT Store that’s coming later this month.
The launch of GPT-4 Turbo —> faster, and 2-3x cheaper than GPT-4, it’s trained up to April 2023, with a context length of 128k tokens (so can fit the equivalent of more than 300 pages of text in a single prompt).
GPT-4 Turbo can also accept images as inputs —> use cases include generating captions, analysing real world images in detail, and reading documents with figures.
The launch of a new text-to-speech (TTS) model and API —> generating voice was the missing modality in OpenAI’s ecosystem.
The launch of Whisper v3 —> the latest version of OpenAI’s cutting-edge Automatic Speech Recognition model features improved performance across languages.
Promises to defend business customers against copyright claims.
🆕 Elon Musk announced Grok, an LLM with a “rebellious streak”, positioned as having the unique advantage of possessing real-time knowledge of the world via the 𝕏 platform. More here.
🇪🇺 Big things also happening in Europe, as German company Aleph Alpha raised $500m - positioned as a European alternative to OpenAI, giving clients “full sovereignty” over the implementation of AI into their businesses
💰 IBM announced a new $500 million enterprise AI venture fund.
🏥 A couple of pretty cool, and potentially highly impactful, healthcare-related developments…
Google DeepMind introduced updates to their groundbreaking AlphaFold model, with a transformative advance in drug development —> the model can now predict the structures of various biomolecules, including nucleic acids and ligands, with near-atomic accuracy. This expansion from proteins to a broader range of molecules could significantly cut drug discovery times and inform early trials.
Canadian researchers have developed an AI model that can detect type 2 diabetes from a person’s voice in less than 10 seconds, when coupling the model with basic health data such as age, gender, height, and weight —> providing a non-invasive, low-cost, quick, and potentially remote, approach to diagnosis.
On safety and regulation ⛑ 👩🏻⚖️
✍️ At the AI Safety Summit, the UK Government published the “Bletchley Declaration”, agreed by 28 countries including the US and China, as well as the European Union, claiming to establish shared agreement and responsibility on the risks, opportunities and a forward process for international collaboration on frontier AI safety and research. More here and here.
🇺🇸 In the US, President Biden issued an executive order to set standards for AI safety and security,
Here’s the fact sheet.
There is of course so pushback —> investors, researchers and founders have submitted a letter to President Biden regarding the AI Executive Order and its potential for restricting open-source AI, stating that the definitions used in the executive order are too broad and will be harmful to academia, open source and smaller companies.
💬 Meanwhile, in the case against stringent regulation…
Andrew NG has commented on how large tech companies are spreading AI fear in order to argue for legislation that would be very damaging to the open-source community. More from Andrew here.
The Economist warns that governments must not rush into policing AI.
If you know someone that would be interested in the contents of this newsletter, please forward it onwards.