The 7 hot topics #Data and #IA of this 6th edition are the solutions for the performing company. What are specifically the trends and topics to track in 2022? Here our videos to find out the answers with images. Discover the new edition of the 7 Data Hot Topics of the Year! The program: Green AI, FinOps, Smart-X, Enterprise Platform… let’s go! Happy New Year everyone 🙂
1. Green AI
AI: problem or solution to the climate emergency?
With the exponential growth of data volume and uses, AI presents a duality regarding environmental and climate issues.
On the one hand, Data and AI are part of the solution to these global problems.
On the other hand, the energy consumption of data centers and AI algorithms is alarming.
To reduce their environmental footprint, organizations must develop green uses while committing to digital sobriety and frugal AI.
Optimizing the cost of Data Platforms in the Cloud
As data platforms are built in the cloud, their pricing model is reshaping traditional financial models.
The CAPEX / OPEX strategy allows great flexibility without requiring a capital budget.
But pay-per-use pricing also means a greater need to control projects and usage.
The FinOps approach should be adopted to get the most out of cloud solutions with a controlled and optimized expense.
3. IoT + AI = Smart-X
The era of industrial data
After years of focusing on personal data, sensors are now driving the exponential growth of global data volumes.
Empowered by AI, sensor data is making objects intelligent and transforming our daily lives.
Smart City 🏙 Smart Industry 🏭 Smart Building 🏢 Smart Agri 🌱… : With the Internet of Things (IoT), entire sectors are being reinvented.
The high-speed deployment of connected objects is revolutionizing uses and marks the advent of the industrial data era.
4. Enterprise Platform
Opening the enterprise to its ecosystem through data and AI
Platform companies increase their value by leveraging data to open up to their ecosystem.
APIs and data marketplaces create direct value by monetizing data.
Open Data and Data Sharing enable collaboration and knowledge sharing.
Federated learning allows AI to be trained from data shared between actors to revolutionize the sectors of Health, Employment, Transport, Environment…
5. Industrialization of AI
The indispensable and fastidious work of data labeling
Supervised learning algorithms must be trained on labeled data.
This labeling can require a significant but imperative effort that slows down AI initiatives.
Dedicated tools, themselves fueled by AI, are available to accelerate and automate this activity.
Labeling is one of the components of MLOps, the approach that brings companies into the industrial phase of AI.
6. Data Mesh
The data decentralization revolution
After years of hyper-centralization of data in warehouses and lakes, an agile and modern alternative is needed.
Organized by business domain, Data Mesh aims for the best balance between centralization and decentralization.
This concept is shaping not only architectures, but also the governance and organization of teams, projects and products.
The new data platforms based on the cloud and data virtualization allow an accelerated deployment of this approach.
7. Trust and Sovereignty
Data and AI at the heart of European regulatory projects
In the aftermath of GDPR, Europe is accelerating its regulatory pace.
Digital Market Act and Digital Services Act to regulate the digital space and control the giga-platforms of GAFAM and others.
Data Act and Data Governance Act to regulate the economy and data exchanges.
AI Act to frame the uses of AI and promote an ethical approach in line with European values. With these projects, Europe is using the regulatory weapon to make its voice heard and try to assert its sovereignty.