Data & AI Newsletter - Winter edition 2023
As the festive season approaches, we're thrilled to reveal the latest batch of AI and Data news!
These months have been particularly exciting for our team, as we recently organized an AI bootcamp that brought together a diverse group of talented participants, and we were thrilled to showcase our work at a healthcare and innovation conference GIANT, emphasizing how AI can be harnessed to enhance healthcare outcomes.
In this newsletter’s edition we compiled a wealth of the most recent news, including the EU's groundbreaking AI Act, Google's latest language model, Gemini, and Amazon Web Services's generous support for INSAIT. And to help you stay up-to-date on the latest learning resources, we've compiled a list of webinars, summit recordings, and free online courses.
Whether you're a seasoned AI practitioner or just starting your journey, we encourage you to immerse yourself in these enriching resources and stay abreast of the latest developments shaping the future of AI and Data Science.
NEWS AND INSIGHTS
•••••••••••••••••••••••••••••••••••••••••••••••
▶️ EU AI Act: first regulation on artificial intelligence
On December 9, 2023, after three days of extensive negotiations, the Council and European Parliament negotiators reached a provisional agreement on the EU AI Act. This legislation marks the first-ever dedicated law on artificial intelligence, establishing a world-leading regulatory framework to ensure the safety, legality, trustworthiness, and respect for fundamental rights within AI systems.
Addressing specific risks, the AI Act categorizes them into four levels, each with tailored rules:
-
Minimal or No Risks: The majority of AI systems with negligible risks can continue without regulation.
-
Limited Risks: AI systems with manageable risks are subject to light transparency obligations to empower users with informed decision-making.
-
High Risks: A broad spectrum of high-risk AI systems will be authorised but with stringent requirements and obligations to access the EU market.
-
Unacceptable Risks: Systems containing deemed unacceptable risks, including cognitive manipulation, predictive policing, emotion recognition in workplaces and schools, social scoring, and certain remote biometric identification systems, will be banned, with limited exceptions.
Gemini is Google’s latest large language model, which the CEO Sundar Pichai first teased at the I/O developer conference in June and is now launching to the public.
Gemini is the result of large-scale collaborative efforts by teams across Google, including our colleagues at Google Research. It was built from the ground up to be multimodal, which means it can generalize and seamlessly understand, operate across and combine different types of information including text, code, audio, image and video.
Developers and enterprise customers will be able to access Gemini Pro through Google Generative AI Studio or Vertex AI in Google Cloud starting on December 13th.
The funds are being made available for demonstrated excellence and growth potential, and are to be used for research in formal methods. The funds will also support salaries for researchers and scholarships for PhD students and students at the institute. Alongside this, a strategic partnership between INSAIT researchers and engineers and Amazon Web Services is also starting.
INSAIT – Institute for Computer Science, Artificial Intelligence and Technology is a non-profit academic research institute located in Sofia, Bulgaria. Founded in 2021, the institute aims to foster excellence in education, research, and innovation in computer science, artificial intelligence, and related fields.
▶️ Nvidia unveils generative AI microservice for accurate answers with enterprise data
The new service, called NeMo Retriever, is part of the Nvidia NeMo cloud-native family of frameworks and tools for building, customizing and deploying generative AI models. It’s designed to provide enterprise organizations the ability to build retrieval-augmented generation capabilities into their generative AI applications.
▶️ Meta and IBM launch ‘AI Alliance’ to promote open-source AI development
Facebook’s parent company, Meta, and IBM launched a new group called the AI Alliance advocating for an “open-science” approach to AI development that puts them at odds with rivals Google, Microsoft and ChatGPT-maker OpenAI.
▶️ The Recursive's Report on the State of AI in CEE
In the midst of 2023's AI revolution, we're thrilled to feature in The Recursive CEE AI Report - an insightful exploration of AI innovation in Central and Eastern Europe.
Dive into the report, discover the transformative power of AI across industries, and join us on this journey as we shape the region's path to becoming a global AI hub. Read more here.
INTERESTING AI STARTUPS AROUND
•••••••••••••••••••••••••••••••••••••••••••••••
🔍 Czech AI Startup Born Digital Secures €2M to Deploy Its Digital Personas Worldwide
The startup builds digital personas to serve as digital public administrator, digital insurance advisor, medical assistant, and many other roles for commercial use. The investment will support their expansion outside of the region. Read more here.
LEARNING RESOURCES
•••••••••••••••••••••••••••••••••••••••••••••••
💡 Webinar recording: AI trends in banking and fintech
The webinar, featuring esteemed speakers Nikolay Nickolovski, MBA, Georgi Nalbantov, PhD, and Svetoslav Ivanov, proved to be a captivating exploration of the transformative potential of AI in the finance industry. Hosted by VectorLabs.AI, the event provided a platform for engaging discussions and the unveiling of innovative strategies. The speakers shared profound insights on AI's impact on banking, effective business strategies, the influential role of data science, real-world applications, emerging trends, and the comprehensive spectrum of AI in the industry.
💡 Summit recordings: Cerebral Valley AI Summit
An exclusive meeting of the top founders and investors in AI at the heart of Cerebral Valley. Find out more about the summit at cerebralvalleysummit.com
💡 Wikipedia + LLM by Stanford University
WikiChat is an experimental chatbot that improves the factuality of large language models by retrieving data from Wikipedia. Large language model (LLM) chatbots like ChatGPT and GPT-4 get things wrong a lot, especially if the information you are looking for is recent ("Tell me about the 2024 Super Bowl.") or about less popular topics ("What are some good movies to watch from [insert your favourite foreign director]?"). WikiChat uses Wikipedia and the following 7-stage pipeline to make sure its responses are factual.
Learn more in the paper here.
💡 [Free course] University of Utah: Local Explanations for Deep Learning Models
This course provides an introduction to local explanations for deep learning models. Local explanations explain why a model makes a particular prediction for a given input. This is in contrast to global explanations, which explain the overall behaviour of a model across a large dataset.
Thanks for reading our monthly digest! Enjoyed this newsletter?
Share it with friends and colleagues who might benefit from it.