Best of LinkedIn: GITEX AI Europe 2026

00:00:00: This episode is provided by Thomas Allgaier and Frennis, based on the most relevant LinkedIn posts about GTEC's AI Europe.

00:00:08: Frennis supports ICT and tech providers with AI ecosystem strategy By delivering independent vendor assessment build versus buy analysis And Ecosystem intelligence that prevents expensive mistakes and positions The provider competitively.

00:00:21: You can find more info in the description.

00:00:24: So uh i think That sets this stage perfectly for what we're going to get into today.

00:00:28: Yeah, absolutely.

00:00:29: Today we're doing a deep dive into the top trends from GTX AI Europe twenty-twenty six and were basing this entirely on the curated insights in most relevant LinkedIn posts that came out of.

00:00:45: We are way past the phase of just marveling at what AI can theoretically do.

00:00:49: Oh,

00:00:50: a hundred percent!

00:00:50: The conversation for ICT and tech professionals has completely moved to the reality of deploying this stuff in an enterprise scale

00:00:56: Exactly...and before you even get into scaling models there's this huge foundational question that everyone was talking about Where does the AI actually live?

00:01:04: And who controls it?

00:01:05: Which brings us to the first major theme we saw everywhere digital sovereignty.

00:01:11: It's funny because historically the European tech sector viewed all this heavy regulation and data governance as a constraint, right?

00:01:19: Like an anchor slowing down innovation.

00:01:21: Yeah especially when you compare it to US or Asian markets but The posts from G-tex show that the narrative has completely flipped.

00:01:30: Governance is actually being wielded as a competitive advantage.

00:01:33: now

00:01:34: I mean, it makes sense if you think about it relying entirely on these massive external data centers controlled by foreign entities.

00:01:40: It's basically like building your company's entire operational future on rented land.

00:01:44: That

00:01:44: is a great way to put in

00:01:46: right.

00:01:46: because of the global provider suddenly changes their pricing or they get caught and some geopolitical crossfire.

00:01:52: Your lights just go out You have no control

00:01:54: And that's why we're seeing such a huge push for sovereign cloud solutions and independent European AI stacks.

00:02:01: Executives are terrified of vendor lock-in.

00:02:03: And invisible

00:02:04: data leaks too!

00:02:05: Yes, exactly so.

00:02:06: when European providers offer this localized infrastructure they aren't just selling software anymore.

00:02:12: They're selling institutional resilience and trust

00:02:15: which is huge.

00:02:16: So if we're securing this infrastructure in building our own power grid?

00:02:20: The next question for a CTO Is well what are actually running on it?

00:02:24: right and based On the market sentiment at the event Nobody is buying Just AI potential any more.

00:02:31: The era of the flashy, generative text demo is completely over.

00:02:36: Totally

00:02:36: over!

00:02:37: The buyers at GITEX were aggressively focused on failure scenarios.

00:02:41: Yeah

00:02:41: they want to know the blast radius Like what happens when a model hallucinates in alive production environment?

00:02:46: The bottleneck really isn't the models capability anymore It's operationalizing it safely.

00:02:51: But wait, let me push back on that a second.

00:02:53: Is the technology actually the bottleneck here or is it us?

00:02:56: I mean looking at the posts.

00:02:57: The real constraint keeping ICT leaders up at night is actually lack of internal human execution capacity.

00:03:03: Oh you mean talent gap.

00:03:04: Yeah and not just needing more prompt engineers.

00:03:07: That's kind of myth.

00:03:10: I see

00:03:12: what you mean.

00:03:14: Like if you want AI to optimize the supply chain, somebody has to physically connect that modern AI model to a twenty-year old legacy ERP system that runs on completely fragmented data.

00:03:26: it's like trying to plug a quantum computer into an nineteen nineties fax machine

00:03:30: which perfectly explains why there was such a massive demand for trusted integration partners at the event.

00:03:36: organizations don't wanna buy another isolated software license then need orchestration

00:03:41: Exactly.

00:03:41: They need partners who can actually map that raw AI capability into their messy corporate reality.

00:03:47: And because human bandwidth is so stretched, enterprises are turning to systems.

00:03:51: they just execute tasks themselves.

00:03:53: Which brings us the huge focus on agentic AI.

00:03:56: Yes agents But we should probably draw a hard line here Because it's super easy to confuse an AI agent with standard IT automation.

00:04:04: Right, if you're listening and thinking this is just a complex zappy or script.

00:04:07: You are dangerously underestimating it.

00:04:09: standard automation is just rules based like If X happens trigger Y

00:04:14: right but what they were showcasing at GI techs Like the supply chain AI agents.

00:04:18: They are dynamic decision systems that evaluate context way probabilities And actively make choices there

00:04:24: basically acting like digital air traffic controllers.

00:04:27: Yeah

00:04:28: and a really practical deployment pattern we saw for this was around ARG co-pilots, retrieval augmented generation

00:04:35: which is essentially how you solve the hallucination problem for these agents right?

00:04:38: Exactly because without our...the language model is just taking a highly educated guess based on the open internet.

00:04:45: but with RAG you force the AI into a closed book test.

00:04:49: It has to search your secure internal database first and generate an answer strictly from that retrieved document.

00:04:55: Makes sense, but taking it a step further if these AI agents are moving beyond internal co-pilots and they're actually acting as buyers or researchers for the company our most product catalogs even readable by an AI

00:05:07: That is million dollar question.

00:05:09: Product data readiness was cited as a mission critical priority for this exact reason.

00:05:15: The whole paradigm is shifting from optimizing for human buyers to optimizing machine buyers.

00:05:20: Wow, yeah think about the B-to-B sales implications there?

00:05:23: I mean if logistics firm's AI agent is scraping web software vendors and your specs are locked inside some beautiful but totally unstructured PDF brochure then agents just gonna skip you.

00:05:35: You could lose a massive deal without a human ever knowing you were even an option just because your data wasn't machine readable.

00:05:42: Exactly, but optimizing for that means opening up huge internal data pipelines

00:05:46: which introduces a terrifying new cybersecurity risk.

00:05:50: I mean, if an agent can execute financial transactions the data and network protecting it have to be absolutely bulletproof.

00:05:56: Which completely tracks with why Data Governance & Cyber Resilience actually overshadowed AI models themselves.

00:06:02: in a lot of LinkedIn discussions

00:06:04: Trust really is the baseline currency here.

00:06:07: We're seeing things like NIS-II compliance and Secure by Design principles becoming massive competitive differentiators.

00:06:14: Oh yeah NIS-II changes the entire development pipeline.

00:06:17: You can't just rush an MVP to market and patch this security later, cryptography and access controls have to be in bedrock of code…

00:06:25: And threat landscape is evolving so fast!

00:06:28: I mean quantum computing moves straight out of theoretical physics labs into executive conversations at GITEX

00:06:34: Because a quantum computer could potentially shatter standard encryption in hours?

00:06:39: Right exactly which makes solutions like Falcon Key Manager so relevant?

00:06:43: People were talking about their launch because they're deploying scalable quantum secure networks today.

00:06:48: Oh,

00:06:48: wow So it's not just a future threat thing.

00:06:50: no It's happening now.

00:06:51: yeah quantum key distribution shifts security from Just math to the actual laws of physics.

00:06:57: if someone tries to intercept a key The quantum state collapses and you know instantly that the line is compromised.

00:07:02: That

00:07:03: is wild.

00:07:03: And we saw that paired with tools like shadow key too which was presented for private data protection.

00:07:08: The focus is entirely on protecting the digital process end-to-end.

00:07:12: Yeah, and speaking of protecting assets there was immense interest in Cosmical at the event Right!

00:07:17: They showcased a remote shielded workspace And that's solution called Endurance right?

00:07:21: Yes A Remote Shielded Workspace Is huge for a distributed European workforce dealing with sensitive defense or health data.

00:07:29: It functionally separates the execution environment from physical device.

00:07:34: So even if an employee's personal laptop gets malware, the enterprise data stays safe in this sovereign enclave.

00:07:40: Exactly!

00:07:41: The malware just can't bridge the gap.

00:07:43: That is brilliant.

00:07:44: We also saw platforms like Pass Machine drawing a lot of focus around secure process digitization because AI agents need verifiable cryptographically secured inputs and AI cannot process a wet ink signature on a scanned piece of paper.

00:07:58: Right,

00:07:58: it needs that digital cryptographic guarantee And that guarantee is what finally allows this intelligence to bleed out at the data centers into the physical

00:08:07: world Which brings us to the next massive frontier discussed on LinkedIn Physical AI Yes

00:08:13: embedding intelligence directly into physical infrastructure, robotics and IoT.

00:08:17: We saw so many concrete use cases for this.

00:08:20: like RoboFinder They introduced a platform that intelligently matches specialist robots to specific healthcare use cases.

00:08:29: So instead of a hospital administrator manually hunting for an automated dispensing robot, this platform maps the hardware capabilities directly to the clinical workflow?

00:08:38: Exactly!

00:08:38: It's an intelligent orchestration layer.

00:08:40: and then there was ThinkCell Assist which launched an AI solution for productivity but with a really specific hook regarding user-owned output.

00:08:48: Oh right addressing the intellectual property nightmare of GenAI

00:08:52: Yeah, making sure the user actually retains the IP rights to what the AI produces.

00:08:56: We also saw Watsonx Automation demonstrating advanced video analytics and Blick showcasing actual driverless car capabilities already deployed in Estonia.

00:09:05: Estonia is such a fascinating sandbox for this stuff.

00:09:08: their regulatory environment for pilots makes them be perfect.

00:09:11: testing ground for edge case physical AI before rolling it out across the rest of Europe.

00:09:15: It really is.

00:09:16: But you know, bringing these massive physical ecosystems to life across Europe requires one fundamental thing.

00:09:24: Capital.

00:09:25: Oh for sure.

00:09:25: the founders on LinkedIn frame this with blunt clarity.

00:09:28: Europe doesn't lack engineering talent or brilliant ideas.

00:09:32: The friction point is commercialization.

00:09:34: Right Founders need better access capital and way less bureaucracy but investors are getting picky.

00:09:41: The winning founder profile isn't just someone with a clever AI wrapper around the generic language model anymore.

00:09:46: No, institutional investors are actively screening for extreme founder resilience and deep tech differentiation.

00:09:54: They want teams solving those unglamorous hard infrastructure problems like the legacy ERP integration we talked about

00:10:02: which is why the physical ecosystem building at GI Techs was viewed as so vital.

00:10:06: You know you might wonder Why A Physical Pavilion Matter So Much For Digital Tech?

00:10:09: But the national delegations basically turned the event into a massive engine for cross-border deal flow.

00:10:14: Yeah, we saw highly structured positioning from The Czech Republic Bulgaria Poland lower Saxony NRW and Dubai

00:10:21: And the consensus was that these curated matchmaking side events provided infinitely more value than just casual networking.

00:10:29: Like, The Czech Pavilion was actively connecting startups with integrators from Germany Lithuania and Poland.

00:10:35: It's

00:10:35: all about securing practical market access.

00:10:38: German companies were actively mapping out international expansion pathways in the UAE...

00:10:43: ...and this physical clustering is only accelerating.

00:10:45: I mean look at the announcement of GTex Italy coming to Rome in twenty-twenty seven

00:10:49: Right.

00:10:50: It just proves that while we're obsessively building digital sovereign infrastructure, physically gathering these ecosystems together is still the ultimate driver of commercialization.

00:11:00: Absolutely!

00:11:01: When you pull all these insights together it's an incredibly complex fast-moving landscape where building sovereign grids deploying autonomous supply chain agents and embedding intelligence into physical robotics...

00:11:13: ...it really brings us right back to this scenario autonomously signing a multi-million euro contract with a vendor's AI in Dubai.

00:11:23: I mean, it leaves you with this provocative thought when the machine autonomously negotiates and signs binding corporate agreement within another machine?

00:11:32: How are our current legal frameworks and sovereign borders going to handle purely machine-to-machine corporate contracts?

00:11:39: Wow!

00:11:40: Yeah... The tech is moving so fast that our legal framework will have to sprint at top speed.

00:11:44: just keep machines on site.

00:11:46: They really.

00:11:47: If you enjoyed this episode, new episodes drop every two weeks.

00:11:50: Also check out our other editions on artificial intelligence, ICT and tech digital products & services cloud sustainability in green ICT defense tech and health Tech.

00:12:00: Thank You so much for joining us On This Deep Dive.

00:12:03: keep exploring stay curious And don't forget to subscribe.

00:12:06: catch you next time.