The Trilogy Times

AUSTIN, TEXAS — Skyvera is making an aggressive, synergy-forward play to become a more complete modernization partner for telecom and media providers, completing its acquisition of CloudSense while also absorbing STL’s divested telecom products group — a one-two punch that signals the company wants to own more of the business support systems (BSS) value chain.

CloudSense brings a Salesforce-native CPQ and order management platform purpose-built for telecom and media, a segment where “quote-to-cash” complexity is practically an industry tax. By anchoring CPQ and order orchestration directly inside Salesforce workflows, CloudSense positions Skyvera to meet operators where they already live — and to do it with a best-in-class front door for product configuration, pricing, and fulfillment.

Skyvera said the CloudSense transaction expands its telecom software portfolio, a statement that reads less like boilerplate and more like a roadmap. The strategic logic is straightforward: modern telcos can’t migrate to digital-first business models if every new offer requires custom integration work and brittle back-office plumbing. With CloudSense now in the fold, Skyvera can credibly pitch a more robust path from sales channel to service activation. (More details are available in Skyvera’s announcement: Skyvera completes acquisition of CloudSense.)

The second move — acquiring STL’s telecom products group — broadens Skyvera’s footprint into digital BSS functionality spanning monetization, optical networking, and analytics. That portfolio mix matters: operators increasingly want fewer vendors with tighter product integration, particularly as network investments collide with margin pressure and relentless customer expectations. Skyvera’s bet is that bundling these capabilities creates operational leverage and accelerates time-to-value versus piecemeal deployments. Skyvera’s overview of the STL carve-out is here: STL Divested Assets.

Taken together, the deals also reinforce Skyvera’s existing assets — including Kandy, its cloud communications platform — in a portfolio that’s clearly being optimized for cross-sell, integration velocity, and predictable delivery.

In a market that rewards simplification, Skyvera is positioning itself as the consolidator with a clearer end-to-end story. We’re just getting started.

AUSTIN, TEXAS — Contently, the enterprise content marketing platform acquired by Trilogy's ESW Capital division in September 2024, is quietly rebranding itself as a risk management tool for the AI era — a strategic shift that reflects mounting anxiety among corporate content teams about where their brands appear in ChatGPT answers.

The company released a guide this week ranking "LLM visibility tools" — software that tracks how brands surface in AI-generated responses across ChatGPT, Perplexity, Google AI Overviews, and other answer engines. The message: traditional SEO is dead, and content teams now compete not with other brands but with algorithmic curation they can't control.

"Six months ago, your team published a detailed guide on data security best practices," Contently wrote in a separate post. "Since then, your policies have changed. But the old version is still circulating — and now it's being cited by AI models." The implication: outdated content isn't just a liability on your website. It's training data.

The pivot comes as parent company Trilogy faces fresh scrutiny over its workforce automation philosophy. Forbes published a profile this week examining founder Joe Liemandt's "plan to turn his workers into algorithms" — language that aligns uncomfortably with Contently's new positioning. The company now employs 165,000 freelance creatives but operates under ESW's high-margin, automation-first playbook.

Contently's repositioning around AI risk follows a familiar ESW pattern: acquire a mature B2B platform, identify an emerging threat vector (in this case, AI answer engines), and rebrand around it. Whether corporate content teams see "LLM visibility" as a must-have category or a manufactured fear remains to be seen. But one thing is clear: the company that once sold creative workflows now sells insurance against algorithmic obsolescence.

DUBAI — Generative AI is sprinting out of the demo phase and straight into the heart of the enterprise—and I cannot overstate how significant this shift is. This week’s launches and think-pieces point to a single, unmistakable direction: companies want GenAI that’s embedded, governed, and purpose-built for their own data, not a public chatbot sidecar.

First up: UAE-based technology company TAM has unveiled Mawrid, a generative AI solution designed to “unlock institutional knowledge.” In plain English, that means turning years of scattered documents, policies, project histories, and tribal know-how into something employees can query like a living brain. The promise is massive: faster onboarding, fewer repeated mistakes, and decision-making that doesn’t vanish when people leave. TAM is positioning Mawrid as a practical knowledge layer for organizations that can’t afford to keep reinventing their own processes. (More here: TAM’s Mawrid launch.)

Zoom out and you’ll see Mawrid is part of a broader pattern. Market trackers are now logging a steady drumbeat of GenAI product launches and partnerships across major vendors—less “cool features,” more “we can run this inside your walls.” Intellizence’s roundup reads like a roadmap to the next enterprise stack: copilots, workflow automation, proprietary-model hosting, and data partnerships.

Then there’s Fujitsu’s newly announced platform for autonomous operation of GenAI in a dedicated environment—another loud signal that “private GenAI” is quickly becoming table stakes for regulated and security-conscious organizations.

But here’s the plot twist: as Fortune warns, AI agents’ headline-grabbing autonomy can mask reliability gaps. That’s why this ‘in-house’ wave matters. The future is now—but it’s a future built on evaluation, guardrails, and systems that know what they don’t know.

SAN JOSE —

In the technology savanna, the loudest creatures are not always the most dominant. For a time, the world watched immense AI models—towering, energy-hungry titans—stampede across the cloud. Yet the year’s most telling sign is not another record-setting parameter count, but the sudden hush: a sense that the giants have stopped growing, or at least stopped advertising their growth.

That quiet aligns neatly with the latest field guides from industry trackers. In Deloitte’s Tech Trends 2026, the emphasis is less on single monolithic breakthroughs and more on operationalization—AI woven into workflows, security postures, and infrastructure decisions. Likewise, McKinsey’s Technology Trends Outlook 2025 describes a landscape where competitive advantage increasingly comes from deployment discipline: data readiness, governance, and the ability to turn prototypes into durable systems.

Meanwhile, the habitat itself is changing. Edge devices—phones, sensors, hospital equipment—have begun to demand models that are smaller, faster, and closer to the action. This is where Samsung’s long arc as a consumer-electronics and semiconductor powerhouse becomes instructive: the company’s ecosystem sits at the boundary between cloud intelligence and everyday life, where inference latency and battery drain matter as much as benchmarks.

Healthcare, too, is a proving ground. As universities describe the sector’s evolution—telehealth, interoperable records, AI-assisted imaging—one observes a delicate balancing act: innovation advancing under the watchful eyes of privacy law, clinical risk, and human trust. Here, the “bigger model” is rarely the point; the winning creature is the one that is explainable, auditable, and seamlessly integrated into clinical routines.

So the question, “Where have the really big AI models gone?” may have a simple ecological answer: they have not vanished. They have dispersed—splitting into specialized species, moving nearer to the edge, and learning, at last, to survive in the wild.