Gartner for IT’s Post

New GenAI foundation models are released every two and a half days, yet nearly half of CIOs report that AI hasn't met ROI expectations. Navigate this dichotomy by determining your organization’s AI pace — and taking the right next steps: https://gtnr.it/3YRiibo #GartnerIT #GenAI #CIO #Technology

  • Infographic titled 'A conceptual framework for sandwich AI' by Gartner, featuring a flowchart with two main sections labeled decentralized and centralized. The decentralized section includes 'Trust, risk, and security' and 'Blended AI' with associated activities like 'Data for centralized AI.' The centralized section is further divided into human-governed operations and tech-driven operations with elements like 'Embedded AI' and 'Commodities of practice.' Color-coded arrows connect the elements demonstrating the flow of AI operations.
Abhinav Mittal ♻️ Governing AI for a Safer World

Responsible AI Advisor | 2X Author | Saved $100m+ in Tech Cost | Technology Thought Leader | 45+ Digital Transformation Projects | IT Strategy & AI Governance | Keynote Speaker | AI Mentor | US EB1A - Green Card Approved

3w

As a responsible AI advisor, while I love the analogy, I’d tweak it a bit! 🍔 Trust, risk, and security aren’t just single layers to stack in the middle — they’re like the sauce that needs to be spread across every part of the AI stack or, better yet, like the Skewer that holds everything together. Without these elements infused throughout, the whole AI “sandwich” risks falling apart, and we end up with gaps and vulnerabilities across the Gen AI value chain. This is a great reminder that building responsible AI is about making trust and safety part of every layer, not just an add-on. 🙏 Here's a healthier recipe for Gartner's AI Burger 😊 .

  • No alternative text description for this image
Maxwell Ngu, B.Sc.

Data Scientist / B.Sc, M.Sc

3w

The AI Technology Sandwich is a practical framework because it acknowledges that successful AI requires not just strong models, but also robust data infrastructure and effective application in real-world settings. It encourages a balanced approach where all layers are given attention, rather than focusing solely on model development. This framework is particularly useful for organizations seeking to operationalize AI effectively, ensuring that each component is in place for sustainable and impactful AI solutions. what are your thoughts, guys????

Rogério C.

CIO | CTO | Diretor Executivo com Experiência Estratégica, Técnica e Operacional

3w

The distinction between 'AI-Steady' and 'AI-Accelerated' approaches accurately reflects the different maturity levels and ambitions we see in the market. I believe the emphasis on behavioral outcomes, alongside business and technology outcomes, is crucial for sustainable AI adoption. In an environment where trust and employee adaptability are essential, strategies like TRiSM prove fundamental in managing risks and fostering a safe culture of innovation. The article reinforces the importance of aligning technology with the specific needs of each sector and shows that AI success goes beyond technology — it involves a careful integration with the organization’s operations and mindset. Congratulations on such insightful content!

Danny Y.

Manufacturing Digital Transformation | Driving Business Growth | Certified Solution Architect | #Hiring

3w

I like the sandwich model. Embedded AI that comes along. Build AI helps create differentiation in business. But I see BYAI (Bring Your Owned AI) should have also been part of Blended AI. Blended AI does exist in both decentralized and centralized form. For AI Platform and Infrastructure for AI, I preferred it to be combined to have closer technology development.

Rafea Al Khaldi

Head of Operations and Key Accounts | Transformational Leadership

3w

Fruitful insights and for sure the sandwich should be digested fast in order to go for a bigger meal....i could guess this would be one of the meanings... 🙂

Shareen J. Harvey

Vice President @ Mechanical Orchard | Marketing I GTM I Customer Strategy I Partner Marketing

3w

It’s also worth asking was/is my infrastructure ready/modernized/ready to optimize and scale AI properly or is there a need to take a step back or reevaluate the core to ensure new technology is integrated correctly and performs as expected?

Wolfram Georg SEYRING, APMP ATO

Managing Director at BidBox International | Business Development Expert | Space | Telecommunication | Defence | Rail | Energy

3w

Insightful

Bert Baeck

Co-founder | CEO | ⚡ Shaping the IoT/OT the data quality/observability category

3w

Excellent overview, Gartner for IT! There is still too much emphasis on AI trends and capabilities, in my opinion. Organizations don’t just need better models; they need better data as well. Doing the flashy AI stuff won’t cut it.

Zied T.

Infra & Cybersecurity Architect | Zero Trust SME CCSP/CISSP

3w

Excellent cheese burger! Just waiting to see if dessert is as well-architected :)

See more comments

To view or add a comment, sign in

Explore topics