CData Software

CData is the data connectivity platform for AI applications, real-time analytics, and enterprise integration—access any data, anywhere. Visit:

[CData Software @cdatasoftware]

Day 1 at The AI Summit London. The team is at booth 313, Tobacco Dock — demoing live, governed access to enterprise data for AI agents. If you're there and working through a data layer problem, stop by!

🔗 https://t.co/DH3zHi6nrb
#TheAISummit

[CData Software @cdatasoftware]

Enterprise AI that queries a copy of your SAP HANA data isn't querying SAP HANA.

CData Connect AI gives Claude governed MCP access to live HANA, query pushdown included.

🔗 https://t.co/sDu6uitMzY

The data doesn't move. The governance doesn't lift.

[David Kleiss @david_kleiss]

71% of enterprise AI projects stall before a single feature ships. The data layer isn't ready.  

@CDataSoftware is at #TheAISummit London this week, booth 313.  

See how MCP-governed access to live enterprise data changes that: https://t.co/xh6cYzmW8H 👀

[CData Software @cdatasoftware]

Databricks named this year's theme 'Build Apps and Agents That Work.' Data access is where most agents break. Stale exports, missing sources, ungoverned access paths.

CData will be at Booth 126, June 15-18.

🔗 https://t.co/dXboFvrqiD
#DataAISummit

[CData Software @cdatasoftware]

This week, the team will be at booth 313, Tobacco Dock, The AI Summit London.

Live demo and QA.
Bring your architecture or your blocker.

🔗 https://t.co/DH3zHi6nrb
#TheAISummit

[CData Software @cdatasoftware]

71% of enterprise AI projects stall before a single feature ships. The data layer wasn't ready.

A thread on why that keeps happening — and what fixing it actually looks like. 👇

#TheAISummit

[CData Software @cdatasoftware]

Governance lives at the connectivity layer — not the output layer. Identity passthrough, OAuth 2.1, least-privilege controls, full audit logging.

The same RBAC rules governing your data automatically apply to every AI query.

[Dr. Bipin Dayal @DayalBipin]

The real edge in AI isn't picking the smartest model. It's engineering the context and tools around it. 

🔗 https://t.co/Yjx6yuuAyY 

Nitin Wagh, CEO at #KariniAI, on where that advantage actually gets built. June 11, noon ET. 

@karini_ai

[CData Software Japan @CDataJapan]

最大97.6%削減！Claudeのトークン使用量を最適化するMCPアーキテクチャ
https://t.co/vwxKs35zbf

[Kazuya Sugimoto @Web API・MCP Holic @sugimomoto]

CData Connect AI のConnect Gateway 機能がHCL Domino にも対応！　オンプレやクローズドな環境のDomino にもセキュアに連携可能に。
#DominoHub 2026 が来週に迫る中、このリリースを今か今かと待ってました！イベントでももちろんお話します！
https://t.co/UpmIQ96jv3

[CData Software @cdatasoftware]

Most teams default to batch bc it's familiar. AI exposes why that default breaks.

https://t.co/4WyoJgKkAL

Agents don't just query data. They write back to source systems. Batch produces read-only copies on a schedule. That's not a latency problem. It's an architecture mismatch.

[CData Software @cdatasoftware]

Everyone's talking about the model. Shawn Sias, VP of Analytics at Elite Media, thinks the real conversation is somewhere else entirely.

His team built it. The results speak for themselves.

Watch this, then register: https://t.co/mgJFPbbhuY
The June 11 roundtable goes deeper.

[CData Software @cdatasoftware]

The answer is a connectivity layer your org controls, sitting between your systems of record and your AI tools, independent of any vendor's policy decisions.

🔗 https://t.co/BsYzNRrAIJ

That's what keeps your AI options open as the stack continues to shift.

[CData Software @cdatasoftware]

SAP's April 2026 API policy restricts third-party AI agents from accessing SAP data directly.

Only SAP-endorsed pathways are permitted.

ODP RFC is blocked starting July 2026.

If your org uses AI against SAP data, the access conditions have changed. 👇

[CData Software @cdatasoftware]

SAP isn't the only vendor doing this. Salesforce has restricted third-party tools from Slack data for AI training. ServiceNow rate-limits non-native agent traffic at scale.

Vendors controlling data are shaping which AI can reason over it.

[Dr. Bipin Dayal @DayalBipin]

The silo problem didn't go away. It moved.  

From data to critical business context fragmented across AI tools, isolated by person, team, and vendor.  

Whitney Myers @wearezuar calls this the context silo. She's joining the roundtable June 11: https://t.co/Yjx6yuuAyY