AI: Striving to Become a Trusted ‘Future Advisor’

Can you imagine what predictive technology looks like? When the foundational capabilities of general large models, the precision of specialized predictive models, the practical value of external tools, and the assurance of trust mechanisms are organically integrated, AI’s insightful new perspective will emerge as a trusted ‘future advisor’ in critical areas such as financial risk control, weather forecasting, public governance, and industrial production. This will provide wisdom support for humanity to grasp future trends and become a significant force in empowering social development and modernizing national governance.

Four Technical Paths for ‘Predicting the Future’

Faced with increasingly complex predictive demands in the real world, researchers have developed two core lines and four specific technical paths around large model predictive technology. These paths are not competitive alternatives but complement each other in different scenarios, together constructing a complete research framework for large model prediction.

The essential difference between the two core lines lies in whether a dedicated model is tailored for the prediction task: one is ‘borrowing a boat to go to sea,’ cleverly utilizing existing mature large language models to complete predictions; the other is ‘building a ship for long voyages,’ reconstructing a dedicated foundational model for prediction. Both paths advance simultaneously, adapting to diverse task requirements.

Directly invoking large language models is the easiest entry point for large model prediction. Researchers translate various predictive tasks into common natural language questions, providing historical information, event backgrounds, and constraints to the model, allowing it to directly assess future trends and output predictions. This approach has a low threshold, requiring no significant modifications to the model; merely changing the use of existing tools can yield impressive results in news event analysis and business trend assessment. However, it falls short in high-precision numerical predictions required in fields like meteorology and finance due to the inherent limitations of large language models in numerical computation and potential factual deviations.

Time series tokenization modeling represents a cross-domain ‘intelligent borrowing.’ It cleverly introduces classic natural language processing ideas into time series data analysis. Through discretization, scaling, and quantization techniques, continuous time series data is transformed into token representations similar to words in language, and training is conducted using architectures similar to language models. The representative model, Chronos, achieves probabilistic predictions and cross-dataset generalization by mapping time series to a fixed vocabulary, significantly reducing development costs. However, this convenience comes with a cost; the data transformation process inevitably leads to the loss of numerical details and quantization errors, akin to a rough polishing of fine parts, which can affect prediction accuracy.

Building dedicated foundational models for time series marks a shift from ‘borrowing strength’ to independent innovation in large model prediction research. Researchers no longer view time series simply as pseudo-text but design pre-training schemes and model architectures tailored to the essential laws of time series data and the core needs of prediction tasks. Google’s TimesFM employs a decoder architecture, showcasing strong zero-shot prediction capabilities; Lag-Llama, developed by multiple universities and research institutions in the U.S., focuses on probabilistic prediction and cross-domain generalization; and Moirai, developed by an American AI company, boldly attempts to adapt to more scenarios through unified training methods. These models are like ‘custom armor’ tailored for prediction tasks, aligning more closely with the characteristics of prediction tasks and achieving higher precision in numerical predictions, making them the top choice for high-precision prediction scenarios.

Reprogramming large language models and multimodal fusion provide a low-cost approach to large model prediction. Research related to Time-LLM confirms that it is possible to involve ‘frozen’ large language models in prediction tasks without retraining massive time series models with hundreds of billions of parameters. By reprogramming to precisely align time series with text prototypes, a feasible pathway for the general large model + specialized adaptation technical route is opened, further promoting the deep joint modeling of text, numerical, and contextual knowledge, allowing predictions to integrate multi-source heterogeneous information like human thinking, better aligning with the complex and variable predictive scenario needs of the real world.

These four technical paths do not have absolute advantages or disadvantages; they are like different keys fitting different locks. When a prediction task requires combining general knowledge and textual background for open-ended trend judgments, the routes related to large language models act as a universal key, providing greater advantages; when tasks pursue high-precision numerical outputs and stable cross-domain generalization capabilities, dedicated foundational models for time series become the custom key for precise matching. They support and achieve each other under different resource conditions and practical task needs, jointly advancing large model predictive technology steadily forward.

Moving Towards Real Application Scenarios

In the research arena of large model predictive technology, international research started earlier and has a more systematic technical framework, delving deeper into foundational research and frontier exploration. Although domestic research started slightly later, it has rapidly caught up with strong momentum, forming unique advantages in scenario adaptation, open-source ecology, and application landing.

International academic research on large model prediction has undergone exciting expansions from text reasoning to multi-dimensional prediction. Early research primarily focused on using large language models for text reasoning and event development judgments, akin to cultivating a small plot of land; in the past two years, it has gradually broken boundaries, expanding into broader fields such as time series, spatiotemporal data, and even scientific prediction, marking the beginning of a new phase of ’expanding territory.’ In the more complex field of scientific prediction, Microsoft’s ClimaX has pioneered the establishment of a foundational model framework for weather and climate tasks, while Aurora, also developed by Microsoft, extends foundational model concepts to the Earth system, capable of simultaneously handling various prediction tasks such as weather, air quality, and wave conditions, akin to equipping the Earth with an intelligent early warning system, showcasing the immense potential of scientific foundational models in complex system predictions.

Notably, the international academic community has maintained a rational and cautious attitude toward the predictive capabilities of large models. Relevant research has found that the excellent performance of large models in standardized tests does not equate to reliability in predicting real-world future events—GPT-4, for instance, performed worse than the median human group in open-world prediction competitions. Around this core issue, international researchers have successively conducted competition research, retrieval enhancement studies, and uncertainty detection research, allowing international research to form a distinct feature of ‘model capability enhancement + prediction result validation + trust mechanism construction’ in equal measure, laying a solid foundation for the practical application of technology.

Domestic research has leveraged the rapid development of general large models to achieve impressive late-stage catch-up, gradually forming a virtuous development pattern of rapid iteration of general large models, systematic review research, and steady progress in application landing. In the arena of building a general model ecosystem, various players have showcased their strengths: Qianwen 3 has developed a complete system for multilingual support and reasoning efficiency optimization, akin to creating a multilingual intelligent bridge; DeepSeek-V3 has achieved technological breakthroughs in high-performance open-source models, making core technologies more accessible; and Wenxin 4.5 has continuously improved in multimodal fusion and engineering deployment, increasingly aligning with practical application needs. Although these general large models are not solely aimed at prediction, they provide a solid capability base for domestic large model prediction research, allowing researchers to stand on the shoulders of ‘giants’ to conduct more targeted studies.

In terms of application landing, the domestic sector is actively exploring ways to take large model predictive technology out of the ‘ivory tower’ and into real application scenarios across various industries. Some studies have deeply integrated expert knowledge with large language models for strategic warning, achieving precise trend judgments and risk identification in complex situations; others have closely combined large models with meteorological monitoring data, attempting to enhance the accuracy and timeliness of short-term precipitation predictions. Although these studies are not entirely equivalent to pure numerical time series predictions, they signify that domestic large model predictive technology is moving from theoretical discussions to practical applications, beginning to explore technology paths that meet local needs and align with industry realities.

Overall, foreign research delves deeper into the development of dedicated foundational models for prediction and scientific prediction, akin to excavating a well-connected tunnel underground, forming a relatively complete technical system; domestic research, on the other hand, is more distinctive in adapting to Chinese scenarios, building low-cost open-source ecosystems, and landing industry applications, akin to constructing high-rise buildings that fit local conditions above ground. With the continuous accumulation of high-quality time series data and industry-specific data in the domestic sector, as well as the gradual improvement of dedicated evaluation systems, there remains significant room for improvement in domestic foundational models aimed at prediction tasks, which will undoubtedly contribute unique and valuable Chinese wisdom to the development of global large model predictive technology.

Bridging the Gap from ‘Powerful to Trustworthy’

Compared to traditional predictive methods, large model predictive technology has achieved a profound transformation from ‘single-point calculation’ to ‘comprehensive judgment,’ evolving from a cold, mechanical computational tool into an intelligent agent capable of understanding context, weighing factors, and providing rational judgments. This unique ability stems from its inherent core advantages, yet it is also like a growing star, steadily evolving towards ’trustworthiness’ and striving to become a reliable ‘future advisor’ for humanity.

The core advantages of large model predictive technology are its innate exceptional capabilities, particularly prominent in practical applications. First, it has strong cross-task transferability. Traditional agricultural yield prediction models cannot be directly applied to stock market trend analysis; switching fields requires starting over. However, large models, with their generalized representation capabilities from large-scale pre-training, can quickly adapt across different fields such as agriculture, finance, and industry with minimal samples. Second, it has great potential for handling complex dependencies. For instance, predicting river water levels during flood seasons is influenced by multiple factors such as rainfall, upstream flooding, and terrain, which traditional models struggle to capture. In contrast, time series foundational models can learn patterns within contextual ranges, akin to having ‘keen eyesight’ to see the connections behind the data. Third, it excels in multi-source information fusion. Traditional meteorological predictions rely solely on numerical monitoring data, while large models can integrate multi-source content such as satellite cloud images, meteorological text reports, and geographic information, transforming predictions from ‘viewing a leopard through a tube’ to ‘panoramic observation.’ Fourth, it has excellent predictive explanation and decision support capabilities. It can not only predict the trend of a particular stock but also explain the influencing factors behind it, such as industry policies and market supply and demand, even providing risk control suggestions, becoming a professional intelligent partner for decision-makers.

Despite its significant advantages, large model predictive technology is not without flaws; there remains a ‘gap’ to cross from the laboratory to real application scenarios. First, the model’s generative and inferential capabilities do not equate to actual predictive abilities. Some models perform excellently in simulated meteorological prediction tests but repeatedly ‘fail’ in actual severe convective weather warnings, simply because the test answers are hidden in the training data, while real predictions require comprehensive judgments on unoccurred events—it’s easy to theorize but challenging to execute. Second, retrieval enhancement addresses symptoms rather than root causes. Although pairing models with information retrieval improves prediction accuracy, it also indicates that models rely solely on their memory of knowledge, akin to guarding an old library, struggling to keep up with real-time changes; acquiring the latest knowledge is crucial. Furthermore, hallucinations and unstable facts pose core obstacles, akin to hidden time bombs. Additionally, constraints related to cost, data, and evaluation systems make large-scale applications challenging. Training high-precision models requires massive computational resources, leading to high development costs; in reality, time series data is fragmented and lacks uniform labeling, making it difficult to produce quality outputs from poor raw materials. Moreover, existing evaluation systems often emphasize numerical errors while neglecting factual stability, causing many models to appear excellent yet struggle to land.

Looking ahead, the development direction of large model predictive technology is clear and focused on ‘bridging the gap from powerful to trustworthy,’ aiming to create a mature technical system that can stably serve real-world decision-making. First, general large models will evolve into dedicated foundational models for prediction, showcasing stronger competitiveness in high-precision demand scenarios like meteorology and finance. Second, tool enhancement will become an important direction, allowing models to autonomously invoke external tools such as search and simulation, akin to equipping intelligent agents with a toolbox to better handle complex scenarios. Third, trustworthiness, controllability, and interpretability will become research priorities; future predictive systems must not only be numerically accurate but also quantify risks and trace judgment bases, which is key for landing in high-risk scenarios. Fourth, accelerating low-cost deployment and industrialization will transform technology from being the exclusive asset of a few institutions into a common tool across various industries as inference costs decrease and open-source ecosystems improve. Fifth, domestic research will deepen localization adaptation, creating dedicated models that combine the Chinese context and local data, ensuring that large models are more accurate, stable, and trustworthy in domestic financial risk control, government warning, and other scenarios.