What’s all this fuss surrounding the AI topic?
Today, embracing Artificial Intelligence is widely seen as vital for progress, innovation, and competitive advantage. While AI isn’t new (dating back to the 1950s), the current boom is fueled by a convergence of factors: exponential computing power growth, massive dataset availability, significant advances in deep learning models (especially transformers like those behind GPT), and democratized tools like TensorFlow and PyTorch.
This combination of capability, accessibility, and demonstrable business impact (automating workflows, enhancing decisions, creating new customer experiences) makes AI literacy increasingly essential. Furthermore, AI leadership is now central to global strategic competition, particularly between U.S. and China, acting as a proxy for economic strength and national security.
Is data really THAT important?
As CTO at Life is Hard (LIH), an insurance tech company connecting insurers and brokers, I’m convinced that without a solid Data Strategy, our AI Strategy will fall short. Many valuable insurance use cases hinge on understanding both historical information and current context (more on specific business cases in a future article).
The intense focus on data is undeniable, illustrated by recent moves: Amazon now automatically sends Alexa voice commands to its cloud [1]; Google’s Gemini uses personal data for customized responses [2]; ChatGPT enhances memory across conversations [3]; and Microsoft’s Recall captures screen activity [4]. These examples, largely reflecting US trends, show an aggressive data acquisition stance, fueling battles between AI developers and content creators [5, 6, 7]. China is likely pursuing similar strategies, albeit less publicly, while the EU prioritizes regulation over rapid tech breakthroughs.
Why this global data appetite? Primarily for competitive advantage through unique insights from proprietary data. Other drivers include accessible AI tools, intense market pressure, demand for hyper-personalization, operational efficiency needs, and risk mitigation.
Technically, data is the fundamental fuel for AI – powering training algorithms, validating models, and shaping insights. Without sufficient, relevant data, AI cannot effectively learn or predict. This massive data processing also raises sustainability questions, especially regarding energy consumption [8] (however, this is an entire topic on its own).
Ok, I’m convinced. What would a data strategy look like?
A data strategy is a comprehensive plan outlining how an organization collects, manages, stores, analyzes, and governs data to achieve goals while meeting regulatory requirements. Key aspects include:
- Data Quality: Ensuring data is clean, standardized, accurate, and monitored is foundational for reliable AI [9].
- Data Governance: Establishing clear policies for data management, ensuring privacy, security, and compliance.
- Data Integration: Seamlessly combining data from often siloed systems to provide AI models with a holistic dataset.
- Data Accessibility: Making data readily available to the right stakeholders (like AI teams) for timely training and analysis.
- Scalability: Designing systems to handle increasing data volumes cost-effectively without sacrificing quality or access.
- Alignment with Business Goals: Ensuring all data practices directly support the organization’s strategic objectives.
Leveraging over 15 years of accumulated insurance data, LIH is evolving how it generates value for clients through data insights. This initiative includes scaling our infrastructure to support new demands and exploring potential data-driven products, all performed in strict adherence to industry regulations and standards like ISO 9001, ISO 27001, GDPR, ASF Norma 4, and DORA.
Final thoughts
The relationship between AI and data is symbiotic: AI learns from data, makes decisions based on data, and improves via data. While AI offers immense potential, it’s intrinsically linked to the quality and strategic management of data. Organizations neglecting their data strategy cannot realize AI’s full benefits – simply put, there is no ‘good AI’ without a robust data strategy. As a final point of consideration, I predict the phrase ‘limitations of human-in-the-loop’ (HITL) will gain prominence. It will be crucial to observe whether this term is used constructively or if it becomes a justification for minimizing necessary human oversight.
References
- New York Post [2025]. Amazon is making a privacy change to Echo https://nypost.com/2025/03/17/tech/amazon-is-making-a-privacy-change-to-echo-that-could-make-device-useless/
- Google [2025]. Gemini gets personal, with tailored help from your Google apps https://blog.google/products/gemini/gemini-personalization
- OpenAI [2025]. Memory and new controls for ChatGPT https://openai.com/index/memory-and-new-controls-for-chatgpt/
- Microsoft [2025]. Privacy and control over your Recall experience https://support.microsoft.com/en-us/windows/privacy-and-control-over-your-recall-experience-d404f672-7647-41e5-886c-a3c59680af15
- TechCrunch [2025]. Jack Dorsey and Elon Musk would like to ‘delete all IP law’ https://techcrunch.com/2025/04/13/jack-dorsey-and-elon-musk-would-like-to-delete-all-ip-law
- OpenAI [2025]. OpenAI asks US for relief from AI rules https://cdn.openai.com/global-affairs/ostp-rfi/ec680b75-d539-4653-b297-8bcf6e5f7686/openai-response-ostp-nsf-rfi-notice-request-for-information-on-the-development-of-an-artificial-intelligence-ai-action-plan.pdf?ref=platformer.news
- Google [2025]. Google joins OpenAI in pushing feds to codify AI training as fair use https://blog.google/outreach-initiatives/public-policy/google-us-ai-action-plan-comments
- MIT Technology Review [2025]: These four charts sum up the state of AI and energy https://www.technologyreview.com/2025/04/17/1115320/four-charts-ai-energy/
- Forbes [2025]: Why 85% Of Your AI Models May Fail https://www.forbes.com/councils/forbestechcouncil/2024/11/15/why-85-of-your-ai-models-may-fail/
Cluj IT will not be liable for any false, inaccurate, inappropriate or incomplete information presented, as the authors are free to choose their approach and relevant topics, within the general guidelines of the newsletter. The opinions expressed by the authors and those providing comments are theirs alone, and do not reflect the opinions of Cluj IT.
Certain links in the articles or comments may lead to external websites. Cluj IT accepts no liability in respect of materials, products or services available on any external website which is not under the control of Cluj IT.
