The real test of investment skill isn't bull market returns, but navigating unforeseen storms. UX Wealth receives monthly allocations from aisot's AI models, which recently identified early warning signals in Berkshire Hathaway, flagging potential issues a month before the company's earnings miss and $3.8 billion Kraft Heinz writedown that surprised analysts. The result: a well-timed exit and re-entry that generated meaningful returns.
Aisot Technologies' AI Insights Platform, demonstrated sophisticated pattern recognition by exiting Berkshire Hathaway shares a month before the conglomerate's disappointing second-quarter results, then strategically re-entering at a lower price.
Our AI models guided UX Wealth to sell the Berkshire Hathaway B position around $485 at the end of June and add it back on at the beginning of August at around $460, capturing a $25-per-share improvement after the company disclosed multiple operational setbacks.
When Berkshire reported in July, the results validated our AI's early warning signals: operating profits declined 4% year-over-year to $11.16 billion in the second quarter, missing expectations, while the company took a $3.8 billion writedown on its underperforming Kraft Heinz stake. Additionally, Berkshire was a net seller of stocks for an 11th quarter in a row, dumping $4.5 billion in equities in the first six months of 2025, while the conglomerate didn't repurchase any stock - disappointing investors who expected buybacks given the stock's correction.
The AI Insights Platform combines traditional forecasting with advanced machine learning to predict investment returns. Our proprietary models analyze market data, economic conditions, and scan millions of news articles, using:
In Berkshire's case, the AI models detected momentum shifts across business segments, early signals from Kraft Heinz board resignations in May, technical distribution patterns, and sentiment changes in news analysis.
The month-long lead time demonstrates our platform's ability to synthesize complex data patterns and detect risks before they become apparent to traditional analysis. Although the main focus of this article is to showcase one particular example on the Berkshire case, our AI models are evaluated on a whole cross-section of stocks and they bring statistically significant contributions to the portfolio construction process. If you are a professional investor who wants to see this in action, book a demo.
AI in action: Exit, wait, re-enter
This trade exemplifies the sophisticated approach to risk management enabled by Aisot Technologies' platform. Rather than permanently avoiding quality companies, our AI models continuously reassess positions based on evolving market conditions. The platform's decision to guide UX Wealth back into Berkshire at $472 shows how our AI Insights Platform can help asset and wealth managers optimize both exit and entry timing while maintaining exposure to fundamentally strong businesses.
While the Berkshire example highlights our AI's stock-level decision making, our AI solution’s most significant value lies in comprehensive portfolio construction. Our machine learning algorithms drive:
Asset allocation optimization - Dynamic adjustments between equities, bonds, and alternatives based on market regime detection
Sector rotation timing - Early identification of industry cycles before traditional analysis catches the shift
Factor exposure management - Continuous optimization of growth, value, momentum, and quality tilts as conditions evolve
Traditional portfolio management often relies on static allocations or quarterly rebalancing. Our AI enables real-time optimization across all portfolio construction levels, from broad asset allocation to individual security selection.
The Berkshire trade demonstrates why we believe our AI Insights Platform provides a sustainable edge in modern portfolio management. Our AI models' ability to process vast datasets, identify subtle pattern changes, and make objective decisions without behavioral biases represents a fundamental advancement over traditional approaches.
Importantly, our platform is designed as an enhancement tool that keeps humans in the loop and in control. Portfolio managers remain the ultimate decision-makers, using our AI insights to inform and improve their investment choices rather than replacing human judgment entirely. This human-AI collaboration combines the best of both worlds: machine processing power with human wisdom and oversight.
As markets become increasingly complex and data-rich, the advantage belongs to systems that can synthesize information at scale while maintaining the flexibility to adapt to changing conditions. This prescient Berkshire trade—UX Wealth avoiding the decline and optimizing re-entry through our platform—illustrates this capability in practice and serves as a compelling return driver for our clients.