As the AI revolution continues, airlines can struggle to develop specific use cases rather than relying on generic models. Jake Bengtson, VP of AI Solutions at Striim, believes that rectifying memory issues is key to upgrading AI applications and achieving more successful results.
The trick right now that everyone is trying to understand is how to get the right context to my models at the right time to make sure you can ground it in the truth that you have within your enterprise.
Everyone has access to the same LLMs at the moment, and our ability to differentiate between competitors will be in bringing together data and adding that as context.
It’s a matter of going through and experimenting with things. I think more and more we’ll see organisations learn how to better evaluate the models for their specific use cases.
- Do you think at the moment the aviation industry, with all the hype around it, is deploying AI for the sake of AI?
- How are legacy systems holding aviation back from deploying AI successfully?
- 3. What needs to change for AI deployment to be impactful to a business and useful to a customer?
Join us at Aviation Festival Asia 2026 to discuss data, technology, and aviation.
For more like this, see:











