SunnyDigest - May 2

Written By Eliana Oviedo

AI is moving fast — not just in what it can do, but where it’s showing up. This week, it’s editing images, aligning with human values, reshaping AI adoption, and transforming enterprise tech.

From front-end features to back-end breakthroughs, here’s what’s new — and what’s next.

GEMINI’S APPLYING FOR A DESIGN JOB

Gemini is stepping beyond text and into the visual world — and it’s not subtle. Google’s latest update lets users circle parts of an image to remove, replace, or regenerate them using natural language prompts.

No sliders. No layers. Just "make this look better," and it does. For creative teams and content marketers, this is a glimpse at a radically faster workflow. For everyone else, it's another sign that AI is quickly moving from novelty to utility.

ANTHROPIC WANTS AI TO GET YOU

As AI systems become more powerful, the question isn't just what they can do — it's whose values they reflect when they do it. Anthropic’s latest research tackles this head-on, exploring how models can be aligned not just to a generic ethical baseline, but to a wide range of deeply personal, even unconventional, value systems.

Their findings show that models can be trained to reason through complex, nuanced belief structures — the kind that don't always show up in training data. The implications are huge: AI that better understands context, respects different worldviews, and adapts to the ethical frameworks of its users.

👉Why it matters: The future of AI won’t be one-size-fits-all. It’ll be systems that flex to fit the people they serve.

😬 AI FOMO IS REAL, AND RISKY

The pressure to “do something with AI” is everywhere. Monte Carlo calls it what it is: AI FOMO — a panic that leads teams to ship GenAI features without clear business cases, skip over data quality, or divert resources from more impactful work.

They argue that this rush often backfires. Teams get distracted. Projects stall. And the core data infrastructure — the thing that actually powers successful AI — is overlooked.

What to do instead: Build from the ground up. Prioritize data quality, reliability, and measurable outcomes. The coolest AI feature in the world won’t work if your data’s a mess.

💬DATABRICKS ANNOUNCES

GA ALERT: SAP BUSINESS DATA CLOUD NOW ON DATABRICKS

While the rest of the industry is busy chasing GenAI dreams, Databricks and SAP are focused on something far more pragmatic: helping enterprise teams use the data they already have.

With their new Business Data Cloud integration now generally available, SAP customers can seamlessly run analytics and machine learning on SAP data directly within Databricks. That’s a big shift from the usual pain of exporting, transforming, and loading legacy data.

The impact: Better access to critical business data, less friction across systems, and faster time to insight — all without rebuilding your data architecture from scratch.

DATA WAREHOUSING IN THE ERA OF AI

Databricks is hosting a webinar on how modern data warehousing is evolving in the AI era — and how enterprise teams can leverage integrations like SAP Business Data Cloud to stay ahead. It’s a must-watch for anyone looking to future-proof their data architecture.

In this post, we explore how businesses can integrate SAP data with Databricks, unlocking the combined power of both platforms. We dive into key components of the SAP ecosystem—such as SAP HANA, S/4HANA, and SAP Business Technology Platform (BTP)—and share effective integration strategies using tools like SparkJDBC and Azure Data Factory. Our goal is to provide actionable insights that empower data professionals to seamlessly connect these powerful platforms and drive smarter, data-driven decision-making.

Eliana Oviedo
Next

SunnyDigest – April 17