Anthropic has accused China of “distilling” its Claude AI model using thousands of fraudulent accounts. China denies cheating. This issue has become a perennial problem, with AI training receiving incessant support from the owners of the source materials used for training. The allegation is the most high-profile yet.
The immediate result was that Alibaba, accused of theft, saw its shares 4.9%. This is a rather expensive example of outrage, if such, and the question of coincidence is entirely unanswerable and unanswerable. Remember that all we really have here are accusations and denials.
Trading volume during the fall was significantly higher than the trading average. Even the suggestion was enough to trigger a market reaction. This is a completely new dimension of problems appearing at the market level and it cannot be the last.
Distillation it’s a longstanding and entirely valid pain point in AI. It is typically applied to training smaller AI models, consumer-grade AI, rather than much larger AI models, which are better used for high-performance big data operations.
Distillation sensitivity has real and potentially dire applications in the marketplace. Training determines the capabilities of AI models. Anthropic recently settled with the claimants regarding the use of source materials. Any question of significant responsibilities will be promptly reviewed by the market.
Anthropic IPOs and a new frontier in market credibility
Meanwhile, Anthropic’s big IPO is already quietly underway. This latest conflict between AI superpowers can get very messy quickly if and when AI performance and market share are affected by issues of distillation. It could also set the stage for a long “round” of AI training models and methodologies.
There’s a lot at stake here, and it’s not just about investors’ money. What the market believes is the real hard currency. The market is buying a lot of AI based on another big thing. It’s not like investors are experts in AI or even know much about it. The market responds to good and bad demand and usually overdoes it.
Anthropic’s IPO has key drivers for valuation. Distillation effectively reduces technological advantages. It invalidates superior AI models and allows less advanced AI a free ride. This is Anthropic’s point of view.
From any competitor’s point of view, marketing a clearly inferior product is suicidal. It is also undeniably stupid to allow their products to depreciate. You can assume that no one in the AI sector wants to be Brand X.
This is the market outlook and will stay with Anthropic as a market leader. Expectations are extremely high and when market expectations are not met, prices fall, sometimes dramatically. It’s likely to be a bumpy ride.
Anthropic’s estimate of expectations is almost too telling
Meanwhile, AI’s entry into the consumer market is a potpourri of macro-scale prospects. Anthropic recently conducted a global survey of “What 81,000 people want from AI.” and the entire spectrum of responses is represented. It is a virtual record of users’ knowledge bases.
This survey deserves some credit as an insight into the reality of Anthropic’s entry into the wider market. People are restless, confused, positive, negative and effectively lost in the whirlwind of new experiences.
Let’s not forget that AI is also not yet fully integrated or accepted in the world.
It’s still a novelty.
It’s still awash with chatbots and agents making headlines that are more annoying than reassuring.
People either need it or they don’t.
They believe it or they don’t.
This is a complicated environment and a moving target for the world’s first major AI IPO. Fear is not a market asset. AI is seen as both a threat and an inevitable part of the future. The fear of missing out has also taken on a whole new meaning. The message is that you don’t have to lose, despite your fear.
At the highest levels of pure functionality, such as Cloud, AI is a huge asset. Its applications are clearly valuable. However, the user’s skills are not at that level. Using AI to read contracts makes sense until it misreads contracts. Reporting depends on external data sources.
Global AI is here
This level of acceptance does not necessarily translate into the consumer market. Whether AI is American, Chinese or European, many common factors will determine the global state of AI in the future.
Special markets: AI will need to specialize and diversify to meet the demands of specialists. This directly affects training, distillation and operational capabilities.
Sciences: The most diverse and demanding markets for AI, the sciences, will have new research and development capabilities. They have to do this, mainly because of the excruciating costs of their work.
Global media, arts and entertainment: The current extremely expensive state of global media production will naturally gravitate towards AI. Even the highly resistant gaming sector is adopting a human+AI form of AI applications for purely practical reasons. The drive for cost and time efficiency makes AI a great asset. Ironically, it can divert money from production to creative resources.
Retail: Online retail can and does benefit financially from simple, almost blockchain-level commerce. At the consumer level, AI may very well be an acquired taste, but it delivers value in terms of point-of-sale communication and issue resolution.
Security: The clamoring demand for better security is one of Anthropic’s big market strengths and will be an ongoing, defining issue for global AI. This is a real arms race and losing is not an option.
The net effect of this inevitable proliferation of AI at all levels of applications is creating a need for common platforms and common ground. Global AI is likely to merge into a range of common practices and adapt accordingly.
The current major earthquake-like effects of AI will need to be put into workable forms simply so that people can get on with their lives. Anthropic’s IPO is the first real global market test.
