Institutional Trust and Operational Complexity Outweigh AI Hype

Investors often overlook the value of institutional trust and operational complexity because they are distracted by the hype surrounding AI. While some argue that AI will make legacy software platforms obsolete, Wix CEO Avishai Abrahami suggests this view ignores the deep trust required to manage enterprise data. The real competitive advantage lies in bridging the gap between theoretical AI capabilities and the messy reality of business operations. Those who understand the difference between prototyping with AI and maintaining a reliable business system will be better at identifying which companies will survive and which are merely hype.

The illusion of vibe coding and the reality of friction

Current market sentiment treats AI as a universal tool that can replace established software. The logic is that if AI can write code, it can replicate any platform. Abrahami disagrees, pointing to the complex business logic embedded in platforms like Wix or Salesforce. Building a simple website is straightforward, but managing delivery schedules, staff, and payroll is a different level of complexity.

You are not going to vibe code Shopify. No matter how good you are, today we are trading on other companies news.

-- Avishai Abrahami

When Abrahami tested whether professional developers could build vertical business logic using AI tools, they failed to achieve the desired result after weeks of effort. This reveals a systems dynamic: the last mile of software, which involves the specific and interconnected needs of a business, is where the real value exists. AI is good at generation, but it struggles with the high stakes reliability required for enterprise operations.

Why the market misprices trust

Systemic trust is a non-transferable asset. Large institutions do not just use Salesforce for its interface; they use it because they trust the platform to secure their data. If a platform becomes a database scraped by AI agents, it loses its function as a trusted intermediary.

The market is currently assigning low value to the core business of companies like Wix while overvaluing AI hype. Abrahami notes that this creates a cycle of mispricing. When the market ignores company fundamentals to chase the latest AI narrative, it creates an opportunity for those who understand that the SaaS apocalypse is a failure of the market to calculate the durability of institutional trust.

The hidden cost of frontier models

Many assume that startups using off-the-shelf frontier models are leaner and faster. Abrahami counters this by highlighting the advantage of training proprietary models. By using Wix data on user intent, they have created a model that is faster, cheaper, and more accurate for their specific use case than generic, high-level models.

We do not need the knowledge that a lot of them have on Chinese poetry because they do, you know? They have been trained on that but we do need the knowledge that I have to focus and increase the knowledge they have.

-- Avishai Abrahami

This is a systems thinking trade-off: generic models have high breadth but low utility for specialized tasks. By investing in proprietary models, Wix is optimizing for the quality of the user experience. Companies that own their data feedback loops will outperform those that rely on generic AI, as the latter will always be limited by the costs and constraints of their providers.

Key action items

• Audit your AI dependency: Distinguish between tasks where AI provides efficiency, such as content generation, and core business logic where reliability is non-negotiable. If you outsource core logic to a generic model, you create a long-term dependency risk.
• Prioritize domain-specific data: Stop trying to build a general AI solution. Focus on the proprietary data your company holds that frontier models do not. Use this to train specialized models that solve specific problems for your users.
• Adopt the CEO sleep strategy: Protect 3 to 4 hours of deep, uninterrupted work time daily. As Abrahami notes, this is the only way to move from fighting fires to solving complex, structural problems that others are too reactive to see.
• Re-evaluate talent retention: Accept that talent turnover is inevitable in volatile markets. Instead of fighting it, focus on maintaining a high-density team culture. The goal is to ensure that those who stay are the highest-impact contributors.
• Shift from growth to quality metrics: As the hype cycle cools, the market will stop rewarding AI-first narratives and start punishing companies with unsustainable AI-agent costs. Prioritize operational efficiency and model quality over rapid, high-cost feature deployment.