Beyond chatbots from DeepSeek to Robotics: AI’s Path Toward Real-World Impact

Wednesday, February 5, 2025

My 93-year-old mother read about President Trump imposing tariffs on Canada and other countries. She asked me, "Why Canada, an ally?" The answer lies in oil—Canada supplies four million barrels per day to the U.S.

According to Reuters, in the week ending January 3, 2025, U.S. imports from Canada hit a record 4.42 million barrels per day. The Congressional Research Service confirms that Canada was the largest supplier of crude oil to the U.S. in 2023, accounting for nearly 60% of total imports.  

On January 27, 2025, I wrote about DeepSeek. Now that the initial hype has settled, the conversation around AI has shifted. Large language models (LLMs) like DeepSeek are no longer seen as the ultimate goal but rather as stepping stones to more impactful applications in robotics, automation, and predictive analytics. Here are a few thoughts:

1. The Diminishing Hype Around LLMs   While models like DeepSeek and GPT-4 remain groundbreaking, the field is moving toward smaller, more efficient AI models that run directly on devices—phones, AR glasses, and more—reducing dependence on cloud computing. That said, LLMs still play a critical role, especially in enterprise AI solutions, ensuring they remain valuable even as their revolutionary appeal fades.

2. Humanoid Robotics: The Next AI Frontier   The next major wave in AI will likely be robotics, with companies like Tesla (Optimus), Figure AI, and numerous Chinese firms investing heavily in this space.  Given China’s dominance in mass manufacturing and AI development, it is well-positioned to capitalize on this shift, especially as aging demographics create demand for eldercare robots.

3. The AI Race and Geopolitical Tensions   China is advancing rapidly in AI adoption, particularly in manufacturing and service industries. However, this acceleration could heighten geopolitical tensions, leading to an AI arms race.

Looking ahead, AI will move beyond chatbots and text generation into real-world applications like robotics and automation. LLMs, while still foundational, may become less revolutionary over time as AI integrates more deeply into everyday life.

The bigger question remains: will global powers collaborate on AI development, or will rivalry shape the next era of tech progress? Unfortunately, for now, competition seems to be prevailing over cooperation—because, get real, after all, we are human.

views
13 responses
Yanwen Xia upvoted this post.
Competition defines humans
From a friend, “ 美好的远景!希望能活到机器人为养老服务的那一天[Grin][Grin]”
My friend and I talked about DeepSeek a few days ago, I told her that chatbots绝不是它的目标,一定会很快进入机器人应用和军事装备上。The race is on!
9 visitors upvoted this post.