CITIC Securities: AI Disrupts the Narrative of U.S. Internet Stocks, Short-term Overinterpretation

China International Capital Corporation (CICC) research report says the AI storyline disrupting U.S. stock internet stocks has been exaggerated in the short term. In consumer scenarios, the incremental AI experience is limited; there are cost constraints on AI substitution; and model companies themselves have capability boundaries. Therefore, AI is more likely to be a cooperative relationship with existing internet platforms rather than a substitute relationship, and some high-quality companies have been clearly wrongly sold off. CICC suggests focusing on companies that have AI-era competitive barriers such as links to the physical world, strong network effects, accumulated data and algorithms, and high-quality content IP, while also looking for directions in which demand expands as AI penetrates further.

Full text as follows

Outlook | U.S. stock internet investment logic under the backdrop of AI Agents

The AI storyline disrupting U.S. stock internet stocks has been exaggerated in the short term. In consumer scenarios, the incremental AI experience is limited; there are cost constraints on AI substitution; and model companies themselves have capability boundaries. Therefore, AI is more likely to be a cooperative relationship with existing internet platforms rather than a substitute relationship, and some high-quality companies have been clearly wrongly sold off. We suggest focusing on companies that have AI-era competitive barriers such as links to the physical world, strong network effects, accumulated data and algorithms, and high-quality content IP, while also finding directions in which demand expands as AI penetration increases.

▍ AI progress is rapid, but it is not all-powerful.

Over the past two years, three major bottlenecks that have constrained the scalable application of Agent-scale models—namely model reliability, inference costs, and ecosystem interoperability—have all seen substantive breakthroughs. This has driven deeper, more productized deployments of AI in core internet scenarios such as e-commerce, advertising, and content. At the same time, market fear that existing internet platforms may be disrupted has continued to grow. We believe AI’s actual impact will be milder than market expectations, with constraints coming from three levels:

1） The incremental AI experience in consumer internet scenarios is overestimated. Unlike the B side, in C-side decision chains such as shopping, transportation, and dining, there is inherently limited repetitive work that AI can significantly improve, and core issues such as information completeness, payment security, and platform reliability have not been effectively resolved under existing AI frameworks;

2） Constraints stemming from the overall cost of AI substitution mean existing platforms cannot be comprehensively rebuilt. After superimposed implicit costs such as subscription fees, maintenance spending, and interface integration, the cost-effectiveness is significantly worse than directly connecting to a mature ecosystem;

3） Capability boundaries and the diseconomies of scale phenomenon of large model companies themselves mean the probability of large model companies cooperating with vertical platforms rather than substituting them is higher. The profitability pressure after listing will also force them to focus on core capabilities. Therefore, we suggest taking an objective and calm view of AI’s impact on the internet sector. What AI brings is structural differentiation rather than systemic disruption.

▍ Finding higher-certainty and structurally incremental opportunities in the AI era.

At the impact level, on the one hand, AI brings quantifiable business increments to internet platforms through three paths: improving user experience, optimizing operational efficiency, and opening up new business scenario. On the other hand, the substitution effects of AI-native products on traffic entry points such as traditional search and content distribution are becoming evident.

At the barrier level, against the backdrop of the above mixed picture, whether platforms can sustain their competitive moats into the AI era is the key to judging investment value. We believe that companies with links to the physical world, strong network effects, accumulated data and algorithms, and high-quality content IP will stand out in the AI wave.

At the opportunity level, through traffic diversion and disintermediation, a considerable portion of commercial value will be captured at the model layer rather than flowing to existing platforms. However, some internet platforms still have opportunities to share in the AI dividend. Their shared characteristic is that they are outside the scope of AI’s direct replacement, but service demand will expand systematically as AI penetration increases. Advertising platforms, streaming media, cloud computing, and so on are typical examples. Based on two major factors—competitive barriers and AI opportunities—we conduct comprehensive quantitative and qualitative analysis. We divide the key covered U.S. internet companies into four quadrants: beneficiaries, points of divergence, safe harbors, and losers, among which some companies are more likely to benefit from this round of the AI wave.

▍ Risk factors:

Risk that AI progress far exceeds expectations and leads to increased impact; risk of overly heavy AI infrastructure investment and uncertainty in returns; risk that AI disrupts the content ecosystem; risk of intensified market competition; risk that regulatory policies related to data and platform operations are tightened, etc.

▍ Investment strategy:

We believe that while AI has achieved breakthroughs in key bottlenecks and accelerated deployment, the narrative that AI will disrupt the internet has been overly pessimistic. This mainly stems from limited incremental AI experience in consumer scenarios, cost constraints on AI substitution, and capability boundaries of large model companies. We suggest focusing on companies that have AI-era competitive barriers such as links to the physical world, strong network effects, accumulated data and algorithms, and high-quality content IP, while also finding directions in which demand expands as AI penetration increases.

（Source: Jiemian News）