2025’s observability predictions
As the use of AI-driven solutions is growing, data regulations are becoming a moving target, and major outages have increased the scrutiny of technology vendor security, observability is transforming from a reactive to a proactive practice. What’s next in 2025? These are the five observability predictions according to Splunk.
Students help solve the lack of talent
While threats are growing, budgets and talent are lagging behind. Universities, public sector, and tech companies will collaborate on student-run Security Operations Centers (SOCs) to address cybersecurity talent shortages, Splunk predicts. This provides hands-on experience for students while offering organizations cost-effective security capabilities. The approach will be particularly relevant in Europe, where the shortage of qualified professionals continues to grow. In addition to the direct benefits, this model also promotes a sustainable inflow of new talent into the cybersecurity sector.
Geopolitics will reshape data residency regulations
The growing wave of nationalism will keep shaping cyber policies, leading to fragmented regional regulations. In response, CISOs, CIOs, and general counsels will collaborate more closely to navigate and address evolving compliance requirements proactively. Organizations must adapt their data management strategies to comply with diverse regional standards, requiring closer collaboration between CISOs, CIOs, and legal teams.
AI will democratize data and provide new opportunities
The rapid rise of AI is revolutionizing how data is collected, analyzed, and utilized. By 2025, AI will extend beyond being a business tool to play a significant role in societal applications, Splunk forecasts. AI will empower organizations to harness data more effectively, fueling innovation in fields like agriculture, healthcare, and conservation.
For instance, AI-powered drones could optimize precision irrigation or monitor forests. However, unlocking AI's full potential requires businesses to prioritize training and upskilling. This investment is crucial to ensure that employees across diverse sectors benefit from technological advancements.
The future of LLMs will be small
Do we need huge LLMs that know everything? If we reduce models to a reasonable size that fits specific use cases, we can reduce costs significantly. Companies will therefore shift from large AI models to domain-specific SLMs, offering lower operational costs, improved accuracy, and reduced environmental impact. Industries like logistics and healthcare will implement SLMs for specialized automation tasks, Splunk predicts.
AI Implementation will be ROI-Driven
While the initial AI hype is starting to wear off, we are seeing a shift toward more practical projects with real targeted returns for businesses, such as boosting the productivity of entry-level workers. AI is able to help lower the barrier to entry and effectiveness for employees who are early in their careers – or are switching jobs.
What’s more, AI adoption will expand beyond business applications to broader societal uses in agriculture, healthcare, and environmental management, Splunk predicts. Organizations must therefore invest in workforce training to fully leverage AI capabilities across all sectors.
