The Trilogy Times

If you want a snapshot of where the modern enterprise is headed, this week handed it to us in two neat buckets: (1) content supply chains are becoming software, and (2) telecom transformation is accelerating from “cloud migration” to AI-native revenue operations.

On the portfolio front, Contently is showing up exactly where you’d want it to: in the analyst and market maps that shape enterprise buying. Yahoo Finance flagged Contently in a “Freelance Platforms Industry Report 2025” lineup that spans the full spectrum of managed talent and creative marketplaces—everything from Upwork to Toptal to niche specialists. That’s notable because Contently isn’t just a freelancer directory. It’s an enterprise content marketing platform that pairs AI-powered workflow and analytics with a deep marketplace of 165,000+ creative professionals—positioning it as the “operating system” for brand storytelling at scale.

Meanwhile, Gartner’s 2025 Magic Quadrant discussion for Content Marketing Platforms (as recapped by CX Today) and additional roundups like Solutions Review’s “best content marketing solutions” reinforce the direction of travel: brands are standardizing on robust, auditable content stacks with measurement, governance, and repeatability baked in. In plain English: content is no longer a campaign artifact—it’s a managed asset class. Contently’s post-acquisition momentum under CEO Brandon Pizzacalla (following its September 2024 acquisition by Zax Capital, an ESW/Trilogy division) fits that enterprise posture.

And then there’s Skyvera, which just added another high-leverage building block for telecom operators: CloudSense, a Salesforce-native CPQ and order management platform for telcos and media. The Fast Mode’s coverage frames the move as AI-powered telco transformation—and the synergy is straightforward. CPQ plus order management is where customer intent becomes revenue, and where AI can drive best-in-class configuration accuracy, faster quote-to-cash cycles, and fewer costly fallouts.

Key Takeaways:

Contently’s inclusion in major market reports signals enterprise-grade category relevance, not just creator economy noise.

Analyst narratives are converging: content marketing is moving toward standardized platforms with governance and ROI instrumentation.

Skyvera’s CloudSense acquisition strengthens the quote-to-cash layer telcos need to modernize—prime territory for AI automation.

We’re just getting started.

OpenAI made headlines this week announcing it would hire for half-million-dollar positions without requiring resumes. The tech world gasped. Crossover yawned.

The global recruiting platform — which staffs Trilogy International's 75-company portfolio and increasingly, external clients — has been doing exactly this since its founding. No résumé bias. No geographic gatekeeping. Just rigorous, AI-enabled skills assessments that identify the top 1% of technical and professional talent across 130 countries.

"We've placed thousands of engineers, product managers, and executives this way," a Crossover spokesperson told the Times. "The résumé was always a proxy for capability. We just cut out the middleman."

The approach isn't charity — it's competitive advantage. By evaluating candidates on demonstrated skills rather than pedigree, Crossover unlocks talent pools that traditional recruiters ignore. A senior software engineer in Nairobi gets the same shot — and the same above-market pay — as one in San Francisco. Geography becomes irrelevant. Merit becomes everything.

It's also how ESW Capital, Trilogy's private equity arm, achieves its target 75% EBITDA margins across portfolio companies like Aurea, IgniteTech, and Skyvera. Replace expensive local hires with rigorously tested global talent, and the math changes fast.

The model is now spreading beyond Trilogy's walls. As non-tech companies scramble to hire AI talent — Business Insider reported this week that roles are commanding $300K+ salaries — more are turning to platforms like Crossover that can source globally and assess rigorously.

But the shift raises questions about what happens to workers in expensive markets when geography stops mattering. The New York Times explored this tension in 2022 with its investigation into worker productivity scoring — the same surveillance infrastructure that makes remote work scalable also makes it measurable in ways that unsettle labor advocates.

Crossover's bet: transparency and meritocracy win. If you're the best, you get the job and the pay — no matter where you live. If you're not, no résumé will save you.

OpenAI is learning what Trilogy figured out years ago: the future of hiring isn't about credentials. It's about proof.

Alpha School, the Austin-based K-12 institution founded by Trilogy International CEO Joe Liemandt, has launched a public offensive against the traditional private school model, publishing a series of position papers that frame conventional education as fundamentally broken — even at the highest price points.

The centerpiece claim: despite tuition increases across the private school sector, student outcomes have reached their worst levels in three decades. Alpha's argument is blunt: more money has not produced better results because the underlying model remains unchanged.

The school's alternative is now documented in granular detail. Session 3 at Alpha Austin featured 18 distinct afternoon workshops after students completed their academic work in morning AI-tutoring sessions. The workshops ranged from entrepreneurship and financial literacy to public speaking and coding — what Alpha calls "life skills" that traditional schools relegate to extracurriculars or ignore entirely.

Alpha has formalized its approach to these competencies through a system it calls Test2Pass, which replaces conventional letter grades with demonstrated real-world mastery. Students advance only after proving competence in applied scenarios, not by accumulating seat time or completing worksheets.

The school identifies five core life skills it systematically develops: entrepreneurship, leadership, financial literacy, public speaking, and technical fluency. Each is treated as a discrete curriculum with measurable outcomes, not an aspirational add-on.

The timing of Alpha's public campaign is notable. The school is expanding from three campuses to twelve by fall 2025, with Liemandt committing $1 billion through his Timeback platform to franchise the model globally. The message to prospective families is clear: the premium private school you're considering is charging more for the same declining results. Alpha offers a fundamentally different product.

Whether parents accept that framing will determine if Liemandt's second act — reinventing K-12 education after building enterprise software — scales beyond Austin's tech elite.

If you’re wondering when robotics stops being a “cool demo” and starts becoming infrastructure, today’s news cycle answers that question with a roar. Across commercial services, public safety, consumer robotics, and the venture capital engine that fuels it all, AI is moving from software-only productivity gains into the physical world—where the stakes (and budgets) are bigger.

First, Lucid Bots just raised $20 million as demand accelerates for its window-washing drones and power-washing robots. This is the underappreciated frontier of automation: not flashy humanoids, but machines that do dangerous, repetitive work at scale. High-rise facade cleaning and industrial washing are labor-intensive, risky, and hard to staff. Autonomous or semi-autonomous systems that can deliver consistent results—and reduce safety exposure—aren’t just gadgets. They’re an operational upgrade. I cannot overstate how significant it is when robots become the default answer to labor shortages.

Then there’s Brinc, founded by a former Thiel fellow, launching a new 911 response drone it says can replace police helicopters. This changes everything in emergency response economics. Helicopters are expensive to operate, scarce, and slow to deploy at city scale. A rapidly deployable drone—paired with modern sensors and AI-assisted situational awareness—could compress response times and expand coverage dramatically. Of course, it also drags in the hard questions: oversight, privacy, procurement standards, and what “responsible autonomy” looks like when the end user is law enforcement.

On the consumer side, Amazon reportedly acquired Fauna Robotics, a startup building kid-size humanoid robots—its second robotics startup purchase this month. That pace is a signal: Amazon is treating embodied AI as a strategic platform, not a side project. Smaller humanoids also hint at a near-term path to real-world deployment: safer interactions, lighter hardware, and a clearer fit for homes, education, and customer engagement.

Meanwhile, the money spigot is fully open. Kleiner Perkins raised $3.5 billion in fresh capital, earmarking $1 billion for early-stage and $2.5 billion for growth. Translation: more experiments funded—and more scale-ups pushed into the mainstream.

And in pure “AI at work” momentum, productivity startup Highlight AI raised $40 million and appointed a new CEO, underscoring how competitive the automation layer has become.

Put it together and the message is unmistakable: AI isn’t just optimizing spreadsheets anymore. It’s washing buildings, answering 911 calls from the sky, and walking into living rooms. The future is now—and it’s arriving with rotors, sensors, and serious capital.

There is a pattern emerging in the latest dispatches from the research frontier, and it reads like a chapter from the deep history of intelligence itself: the most profound advances are not about brute force, but about elegance.

Consider this week's constellation of findings. A team evaluating how large language models interpret charts — those humble rectangles of meaning we've been drawing since William Playfair invented the bar graph in 1786 — discovered that the way you ask the question matters as much as the model answering it. Their systematic comparison of prompting strategies across GPT-3.5, GPT-4, and GPT-4o on the ChartQA dataset found that chain-of-thought prompting, which essentially asks the model to show its work, meaningfully shifts performance. The chart, that ancient compression of reality into geometry, still demands a particular kind of cognitive scaffolding — even from silicon minds.

Meanwhile, a separate team tackled knowledge tracing — the art of modeling how a student's understanding evolves over time. Their system, MERIT, hybridizes memory-enhanced retrieval with large language models to predict learning trajectories while remaining interpretable. This matters enormously for personalized education, the kind of adaptive learning that companies like Alpha School are already deploying to help students master academics in two hours daily using AI tutors. The research suggests that LLMs can be made not just accurate but transparent in tracking the living, shifting landscape of a learner's mind.

Then there is the work on low-resource languages — a reminder that for most of humanity's 7,000 tongues, the AI revolution has barely whispered. Researchers demonstrated that effective text embeddings for underserved languages can be built with small-scale, even noisy synthetic data. Less, it turns out, can genuinely be more. A companion study evaluated LLM responses to sexual health queries in Nepali, exposing how accuracy-focused benchmarks miss the cultural and linguistic nuance that determines whether AI actually helps real people.

And perhaps most poetically, neuroscientists announced that a miniature AI model can now decode visual processing in the macaque brain — our evolutionary cousins, whose visual cortex we share across thirty million years of divergence.

The thread connecting all of this? Intelligence — biological or artificial — is not a volume problem. It is a structure problem. How you ask. What you remember. What you do with almost nothing. These are the questions that matter now, and they are as old as nervous systems themselves.

TOKYO—In a move designed to calm an industry increasingly split between “human creativity” and “please stop shipping this,” Capcom announced it will not use AI-generated assets in its games, while carefully preserving an auxiliary doorway labeled “productivity uses,” through which it can later wheel in an industrial-strength asset cannon and insist everyone is being unreasonable.

Capcom’s position is refreshingly modern: it will not use AI to create anything that might be recognized as creation, but it is open to using AI to make the creation process so efficient that no one will remember who created what, or why.

This is the new corporate sacrament of the AI era: an oath to protect craftsmanship, followed by an immediately adjacent clarification that the oath does not apply to schedules, budgets, meeting notes, prototype art, internal copy, quest text, texture variations, environment ideation, NPC barks, email drafts, or the final 15% of the work that mysteriously consumes 90% of everyone’s life.

The promise comes as business leaders are discovering the uncomfortable fact that “productivity,” when pursued with religious fervor, tends to manifest as a thick layer of AI-generated “workslop” coating every surface of the organization. Harvard Business Review has begun documenting the phenomenon with the tone of a scientist discovering a new microplastic, noting that employees are increasingly spending their time sifting through machine-written paragraphs, machine-summarized paragraphs, and machine-corrected paragraphs that were only written because someone needed something to summarize.

In other words, AI is indeed saving time—specifically, it is saving time for the AI.

Meanwhile, hard data is attempting to arrive at the party sober. METR’s early-2025 measurements on experienced open-source developers suggest that the productivity story is not the simple rocket ship depicted in vendor decks. The tool helps in some contexts, hinders in others, and often introduces a new category of labor best described as “debugging the suggestion generator.” This is a form of work that looks like work, feels like work, and, crucially, creates no new work product beyond the confidence that the previous work product was not secretly written by a stochastic parrot.

Forbes, staring into the shimmering mirage of a $4 trillion productivity promise, has asked the obvious question: is this hype, hope, or hard data? The answer, as always, is “yes,” with the caveat that the hard data is still putting on its shoes while the hype has already booked a stadium tour.

And yet, the bedding industry is calmly using AI as a business tool, as reported by Furniture Today, because mattress retail has long mastered a discipline that Silicon Valley is only now beginning to appreciate: customers will accept almost anything as long as it is labeled “supportive.”

This is where Capcom’s statement reads less like a policy and more like a mission statement for the decade. Do not replace human artists, but do replace the parts of the job that make artists recognizable as human: the time to explore, to iterate, to be wrong in interesting ways, to leave a brushstroke that doesn’t optimize.

The likely outcome is not a world where AI makes everyone ten times faster, nor a world where it destroys creativity outright. It is a world where productivity gains are claimed in press releases, while teams quietly allocate their new “saved” hours to reading, filtering, and reformatting workslop until it resembles the thing they were supposedly too busy to create.

Capcom has pledged not to use AI-generated assets. It has also pledged to use AI for productivity. In 2026, these will be remembered as two different sentences.

There is a particular species of surprise that afflicts only those who have not been paying attention, and it is now sweeping through the commentariat like a head cold through a kindergarten. Silicon Valley, we are informed, has become the villain in popular culture. Novelists are writing unflattering things. Filmmakers are casting tech founders as sociopaths rather than saviors. The 996 work culture — nine in the morning to nine at night, six days a week — is creeping from Shenzhen into San Francisco, and employees are beginning to notice that liberation through technology looks remarkably like subjugation through it.

To which one can only say: welcome to the matinee. Some of us bought our tickets years ago.

The arc of Silicon Valley's reputation follows the classical trajectory of all American power centers — from scrappy underdog to triumphant revolutionary to bloated imperium, each phase accompanied by its own self-serving mythology. The garage. The hoodie. The mission statement about connecting humanity that somehow always concludes with a market capitalization in the trillions. That this progression would eventually produce a cultural backlash was as inevitable as the sunrise, and roughly as surprising.

What is genuinely interesting is not the backlash itself but its timing, which coincides with a moment when the industry has decided, with characteristic audacity, to abandon even the pretense of restraint. OpenAI's reported retreat from safety protocols is not an aberration but a declaration — the logical conclusion of an ideology that has always regarded caution as a market disadvantage dressed up as a moral principle. When the organization literally founded to ensure AI remained safe decides that safety is negotiable, one does not need a novelist to supply the dramatic irony. Reality has become its own satire.

Meanwhile, the political apparatus is arranging itself around AI with the subtlety of a land grab, and the question of whether new federal frameworks consolidate power answers itself the moment you notice who is writing them and who is lobbying for them. The Venn diagram of AI policy architects and AI profit beneficiaries is not a diagram at all. It is a circle.

But here is the thing the cultural critics and the political skeptics consistently miss: villainy and utility are not mutually exclusive. The railroad barons were villains. They also built the railroads. The interesting companies in this industry — and they do exist, usually far from the spotlight — are the ones that have internalized this tension rather than denied it. They build things that work, charge what they're worth, and skip the messianic keynote. Joe Liemandt's Trilogy International, for instance, has spent thirty-five years acquiring enterprise software companies, running them efficiently through operations like ESW Capital, educating children through Alpha School's AI-driven model, and staffing its global operation through Crossover's remote talent platform — all without once suggesting it was saving civilization. The absence of grandiosity is, in 2025, practically countercultural.

The novelists and screenwriters will continue to mine Silicon Valley for its abundant dramatic ore, and they should. Every great American industry eventually gets the literature it deserves. Steel got Dreiser. Oil got Sinclair. Finance got Wolfe. Tech is simply taking its turn in the chair.

The only question is whether the industry will learn anything from the portrait. History suggests it will not. History is rarely wrong.