Posted on

Meta × Manus: The misread AI deal

Most people read Meta’s acquisition of Manus as another step in the AI agent arms race.

Yes and no.

From a VC lens, this was not a bet on intelligence.

It was a bet on execution scar tissue — something that can’t be rushed, simulated, or cheaply rebuilt.

This was never about “the best model”

Meta already has:

  • Strong foundation models
  • Massive global distribution
  • hardware endpoints (mobile, VR, wearables)

What Meta does not have the luxury of doing is learning from execution failures publicly across billions of user interactions.

Manus had already done that.

The real question isn’t “why agents?”

It’s “why Manus?”

Here’s the non-obvious answer:

Manus crossed a path-dependent threshold where execution reliability — not reasoning quality — became the moat.

Once a company reaches that point, the ‘build vs. buy’ debate stops being a technical decision and becomes a time-risk and reputational-risk decision.

What Manus learned that others can’t shortcut

Most AI agents work in controlled environments:

clean prompts, trained users, bounded workflows, human-in-the-loop recovery.

Manus appears to have learned how agents behave in hostile, real-world environments — the kind Meta operates in.

Also Read: How SMEs can compete like big corporations with the right financial intelligence platform

Three lessons matter:

  • Failure recovery matters more than first-pass intelligence: Real users are ambiguous. Tools break. Instructions are incomplete. Manus learned how to recover without hallucinating or escalating to humans.
  • Long-horizon execution is harder than reasoning: Execution requires memory, intent persistence, and recovery across sessions — where most agent demos collapse.
  • Trust collapses faster than models improve: In consumer platforms, silent failure isn’t bad UX — it’s a trust breach.

Manus learned how to fail visibly, explain minimally, and recover credibly.

None of this is benchmarkable.

All of it is learned the hard way.

Why the acquisition was inevitable

Meta could rebuild these capabilities.

What it couldn’t afford was:

  • Relearning failure inside WhatsApp, Instagram, or wearables
  • Exposing billions of users to that learning curve
  • Absorbing the reputational risk of agents behaving badly at scale

So the real decision wasn’t “can we build this?”

It was “Can we afford to relearn this?”

The answer was no.

The signal for founders and investors

General-purpose AI agents are now a platform game.

Venture-backable paths narrow to:

  • Deep vertical agents with real domain lock-in
  • Infrastructure layers (orchestration, observability, compliance)
  • Acquisition-grade teams with real execution scars

The era is shifting from model competition to execution control.

And the hardest asset to replicate isn’t intelligence — it’s the accumulated cost of being wrong in the real world.

That’s what Meta bought.

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic.

Enjoyed this read? Don’t miss out on the next insight. Join our WhatsApp channel for real-time drops.

Header image generated using AI.

The post Meta × Manus: The misread AI deal appeared first on e27.

Posted on

Value creation: The private equity execution paradox

Private equity has a religion: operational excellence through systematic value creation. The data appear unassailable. McKinsey reports that operationally-focused GPs generate IRRs two to three percentage points higher than peers. Bain finds that structured value creation delivers 3.0x returns versus a 1.9x industry average—a 58 per cent performance premium.

Here’s the heresy nobody wants to admit: the same discipline that creates outperformance is now destroying more value than it generates.

The numbers behind the orthodoxy’s failure

Simon-Kucher’s 2025 study reveals what consulting firms won’t tell clients: two-thirds of private equity (PE) value creation initiatives fail to deliver expected outcomes. One in three business improvement programs produces zero measurable return. In value-destructive deals, more than 10 per cent of employees depart immediately post-close, with the worst transactions losing 21-30 per cent of key talent.

Most damningly: 75 per cent of portfolio company CEOs are replaced within two years—not because they’re incompetent, but because they resist the new owner’s systematic playbook. When AlixPartners surveyed PE practitioners, 75 per cent reported direct experience with portfolio failures caused by CEO-investor misalignment. Yet only 13 per cent conduct cultural evaluation during diligence.

The industry identifies the problem, then systematically engineers the conditions that produce it.

What actually kills value

Simon-Kucher dissected why value creation initiatives fail:

  • Poor implementation: 53 per cent
  • Unrealistic business cases: 37 per cent
  • Portfolio company resistance: 35 per cent

Notice what’s absent? Insufficient KPIs. Inadequate governance. Too little systematisation. The failure mode isn’t under-management—it’s over-management imposed before organisations can absorb it.

I’ve watched this pattern destroy dozens of companies across Southeast Asia. A founder-led B2B software company generating 40 per cent annual growth gets acquired by a PE firm deploying its “proven playbook.” Within six months, board decks balloon from 15 to 60 slides. Hiring approvals stretch from days to three weeks. The product roadmap freezes pending “strategic review.”

Twelve months later, revenue growth has halved, the CTO has quit, and the PE firm convenes an urgent off-site to diagnose “execution challenges.” The playbook wasn’t wrong. The timing destroyed the business.

Also Read: The culture conundrum: Why private equity’s best CEOs still fail and how Moneyball thinking can fix it

The elite firm counter-strategy

Here’s what separates genuine 3x performers from systematisation zealots: they treat frameworks as tools to amplify momentum, not replace it. They recognise that premature systematisation in high-growth companies is value destruction wearing the costume of best practice.

Top-quartile firms do something radically different in year one. They watch. They resource. They remove obstacles. They preserve the operational momentum that justified the multiple acquisitions. Only after stability is achieved do they introduce frameworks—selectively, where they enhance rather than constrain performance.

Bain’s research inadvertently proves this. The 3.0x performers engaging in “structured value creation” aren’t imposing rigid KPI frameworks. They’re making surgical interventions: replacing one genuinely inadequate executive, funding a capital-constrained growth channel, and implementing pricing discipline where none existed. These aren’t cookie-cutter implementations—they’re strategic decisions executed with restraint.

Contrast this with 1.9x performers who arrive with 100-day plans and systematic frameworks deployed identically across portfolio companies regardless of context. Their religion is a process. Their blind spot: process imposed before momentum is established kills the growth they acquired.

“At the end of the day, it’s people and culture that decide whether a system succeeds or fails.”

The advent international lesson everyone misses

The industry loves citing Advent’s ownership of BSV, which doubled revenue growth to 20 per cent annually and expanded EBITDA margins from 20 per cent to 30 per cent. What case studies omit: Advent didn’t impose comprehensive frameworks on day one. They made two interventions—international expansion and pricing optimisation—then resourced them aggressively while leaving the operational engine intact.

This is surgical execution, not systematic transformation. The discipline came from knowing what not to systematise—preserving the sales culture and product velocity that created value in the first place.

Why this matters now

Private equity faces its harshest environment in 15 years: US$3.6 trillion in unrealised value across 29,000 unsold companies, distributions at historic lows, and acquisition multiples at 12x EBITDA. In this context, the industry’s reflexive answer has been more systematisation—more frameworks, more governance, deployed earlier and more uniformly.

The data says this orthodoxy is failing. CEO turnover approaches 75 per cent within two years. Performance gaps between top-quartile and median funds continue widening—not because median managers lack frameworks, but because they’ve mistaken process for performance.

Also Read: Why private credit is becoming the hottest alternative for smart investors

McKinsey’s organisational alignment research remains valid: culture explains 58.6 per cent of variance in execution outcomes. But here’s the inversion consultants won’t acknowledge: you cannot systematise culture into existence. Culture precedes systems, not vice versa.

The firms generating genuine alpha have learned what the rest refuse to accept: systematic value creation is timing-dependent, not universally applicable. Deploy frameworks too early, and you destroy growth. Deploy them when leadership is stable, baselines are established, and organisations have absorption capacity—and systems amplify performance.

The uncomfortable truth

As Bain observes, the cost of market-beating returns continues rising as fees compress. Winners won’t be those with the most sophisticated frameworks but those with judgment to know when frameworks enable versus suffocate performance.

The industry sold the world on systematic value creation. The uncomfortable truth is that the system itself has become the primary destroyer of value. The competitive advantage has become a vulnerability.

Elite firms have discovered that the hardest discipline isn’t imposing rigour—it’s having the restraint to preserve entrepreneurial velocity when every instinct says to systematise faster. That judgment, unglamorous and maddeningly contingent, is now the true source of private equity alpha.

The beatings will continue until morale improves. Or until the industry learns that its most sacred principle requires the one thing PE hates most: patience.

This article is part of David Kim’s Value Creation column. It sits alongside the Asia Value Creation Awards, which aim to recognise PE and VC teams driving long-term, fundamentals-led value creation across the region.

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic.

Enjoyed this read? Don’t miss out on the next insight. Join our WhatsApp channel for real-time drops.

Image courtesy of the author.

The post Value creation: The private equity execution paradox appeared first on e27.

Posted on

Is AI making us lonely? Or is it helping us feel less alone?

We talk about technology as if it separates us. Phones are replacing conversations. Screens replacing faces. AI is replacing humans.

But maybe the real question is not whether AI makes us lonely. Maybe it is how we use it to feel less alone.

The quiet kind of loneliness

Loneliness today does not always look like being alone. It can happen in crowded rooms, busy offices and even online communities. We scroll, we watch, we like, but we do not always feel seen.

That is why connection has become the new happiness. And that is where AI, surprisingly, can help. Not by pretending to be human, but by helping humans rediscover each other.

When AI becomes a mirror

I have watched many learners, especially midlifers, use AI for the first time. At first, they are shy. They talk to it softly, like a stranger they are not sure they can trust.

Then something shifts. They begin to write. They begin to share. They begin to remember. They tell stories they had forgotten,
describe feelings they had never spoken about, and rediscover parts of themselves they thought were lost.

It is not because AI understands them perfectly. It is because AI listens without judgment. That kind of listening gives people courage to speak again, and that is where healing begins.

Connection through creation

A dear friend once told me about a project she started with her ageing mother. Her mother had always dreamed of running a small shop, but life, family and time never allowed it.

When her health began to fade, my friend helped her create that dream online. They built a simple e-shop together using free digital tools. They took photos, wrote short product descriptions and posted them on social media. Soon, friends began to notice, comment and buy small items.

Also Read: Why Singapore startups are sleeping on their secret weapon (spoiler: it’s not AI)

It was not powered by complex AI systems. It was powered by love, curiosity and technology made simple by AI in the background. The shop gave her mother a sense of purpose. It gave her daughter a memory she will never forget.

For a short while, they lived their mother’s dream together. That is what connection through creation truly means. Technology is not only about efficiency. It is about giving people one more way to feel alive, seen and connected.

AI and emotional literacy

AI cannot replace empathy, but it can remind us how to practice it. It can help us reflect, write and reach out. It can turn thoughts into voice, ideas into visuals and memories into legacies.

Used with intention, it becomes a tool for emotional literacy, a way for people to understand what they feel and express it safely.

When I see participants use AI to share stories about their lives, they often say, I did not know I still had so much inside me.
That is not technology at work. That is humanity reawakened.

The gentle reminder

We were never meant to compete with machines. We were meant to grow with them.

AI can help us create, express and connect, but it is still our emotions that give it meaning.

So do not fear it. Use it to find your voice, your joy and your people because the future of happiness is not artificial. It is amplified.

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic.

Enjoyed this read? Don’t miss out on the next insight. Join our WhatsApp channel for real-time drops.

Image credit: Canva

The post Is AI making us lonely? Or is it helping us feel less alone? appeared first on e27.