NVIDIA is indeed powerful, and Powell didn't flip the table, but the debate over whether AI is still viable has become more intense.

Wednesday's NVIDIA Earnings Report Shows that NVIDIA's second-quarter revenue doubled to $13.5 billion YoY, with data center revenue reaching a record $10.32 billion, more than doubling MoM.

Subsequently, Federal Reserve Chairman Powell delivered a speech at the Jackson Hole Global Central Bank Annual Meeting. Although Powell's attitude was firm, he did not say anything surprising - the Federal Reserve will continue to rely on data while waiting for inflation to fall to 2%.

NVIDIA's impressive earnings report and Powell's cautious speech finally allowed the US stock market to safely pass through a long and difficult week. Overnight, the S&P 500 ended a three-week decline, with NVIDIA rising more than 6% this week, marking two consecutive weeks of significant gains.

However, this does not mean that the market can rest easy in the future. Investors are still selling the rips (shorting falling assets during brief rebounds after overselling) instead of buying on dips. This has been the case since August.

At the same time, the main driving force behind the strong rebound of the US stock market this year, AI, has shown signs of cooling off. After NVIDIA's earnings report, the most important report in the AI industry, made a nearly perfect appearance, the performance of the US stock market has been quite strange. On Thursday, the Nasdaq fell nearly 2%, while NVIDIA only rose 0.1%. On Friday, NVIDIA fell 2.4%, while the Nasdaq rose 0.9%.

Since the beginning of this year, the Nasdaq has risen by more than 30%, and NVIDIA has skyrocketed by 220%. The huge gains have made analysts increasingly skeptical about whether AI concept stocks still have room to rise. The debate about the future of AI is intensifying.

Affirmative: NVIDIA Confirms the Promising Future of AI

Investors with positive views believe that even if the stock price seems frighteningly high, there is still room for growth in the AI sector.

NVIDIA's earnings report shows that its non-GAAP net profit in the second quarter increased by 422% YoY, reaching an incredible $6.74 billion, more than any other company in the AI field.

Quincy Krosby, Chief Global Strategist at LPL Financial, said:

NVIDIA confirms the rationality of the future (rise) of AI. This is not a recurrence of the tech bubble.

Currently, many technology companies still need time to monetize AI. Bulls say there are signs that internet companies (the main buyers of NVIDIA chips) are looking for ways to leverage AI technology to advance their own businesses. "You will start to see companies like Microsoft and Salesforce demonstrate their revenue potential in the AI field in the next two to four quarters," said Angelo Zino, Senior Equity Analyst at independent investment research firm CFRA.

After NVIDIA released its earnings report, Wall Street analysts collectively raised their price targets for NVIDIA. According to FactSet, NVIDIA's current target price is $640.71, a 39.2% increase from Friday's closing price of $460.18.

However, some investors have issued warnings that if NVIDIA encounters difficulties in the coming quarters, its high valuation leaves little room for investors to maneuver.

Data from FactSet shows that NVIDIA's trailing twelve-month P/E ratio is 35 times. Bill Fleckenstein, President of Fleckenstein Capital, stated:

If you want to make this kind of bet, you have to maintain a fairly optimistic view of the market and the overall economy.

The strong demand has given NVIDIA the confidence to raise prices, with the company's gross margin reaching 70.1% in the second quarter, far exceeding the 43.5% in the same period last year. Some investors warn that once new competition emerges, NVIDIA's ability to command high prices may not be sustainable.

Fred Hickey, editor of "The High-Tech Strategist," said:

There's no doubt it's a great story, and there's no doubt it's a great number. Will this situation continue over time? History suggests it's unlikely.

Counterpoint: AI may never generate revenue

However, so far, besides NVIDIA, no tech company has made big money from AI. What if generative AI is ultimately proven to be just a "dud"?

An article by Gary Marcus mentioned by Wall Street CN stated that as investors realize the poor performance and lack of "killer" commercial applications of generative AI, we may see the possibility of a "massive and painful adjustment" in valuations.

AI has not yet generated revenue, and it may never do so.

Marcus is the co-founder of the Center for the Advancement of Trustworthy AI and testified before the U.S. Congress this year. He has long been skeptical of the intelligence of neural network models underlying chatbots like ChatGPT. As an example of the unreliability of the model itself, one of the fatal flaws of artificial intelligence technology is that it can create illusions and spout nonsense with a straight face.

A greater concern is that the content generated by generative AI is polluting the datasets for future system training, posing a threat to what some call "model collapse." Generative AI adds more imperfect information and intentionally false information to knowledge bases, further contributing to the "enshittification" (garbageification) of the internet. This means that trained AI will output more nonsense.