实际上,在会议上,Broadcom CEO 是在为前沿实验室的估值辩护,并对 AI 基础设施需求发表了谨慎乐观的看法。
他只是将利率上升列为潜在制约因素,而更高的营收或更广泛的应用可以克服这一阻力。
断章取义地摘选言论,扭曲了他的真实表达。
简而言之:
- AI 半导体需求庞大且持续高企,半导体供应不足以满足市场需求。高投入至少还会持续 2 年
- OpenAI 的估值站得住脚;那些因为对公司抱有极端预期而卖出的人,就像赌场赌徒
- 陈福阳表示,市场误读了 OpenAI 的年化经常性收入,因为其基本面依然强劲
- 如果前沿实验室实现广泛应用并获得可观营收,其融资能力将得到增强,足以压过利率上升带来的阻力
$AVGO CEO 的许多言论其实非常讲究分寸,却被西方媒体的报道扭曲了。
💬 引用 Serenity(@aleabitoreddit)
看到韩国市场的 $SKHY / 三星跌成这样,真让人难过。
还得特别“感谢”$AVGO CEO 火上浇油……
他说 AI 投资者都是赌徒,还称企业级 AI 没能带来投资回报。https://t.co/fHzsD9Na4s
原帖:https://x.com/aleabitoreddit/status/2109852646484201736 Serenity (@aleabitoreddit) on X Wow this is sad to see Korean markets with $SKHY / Samsung crash like this.
原文: So spent some more time to look at more DBS conference material:
- "Two semiconductor giants, $AVGO and $INTC, both stated that breakthroughs in AI will keep demand for chip investment elevated for at least another 2 years. As current supply is insufficient to meet market demand."
- "Current chip production capacity is already far from sufficient to meet demand, and $AVGO CEO expects the shortage to become even more severe in the future." (HKEJ)
- "Regarding OpenAI, Hock Tan stated that the market had misinterpreted the company because its underlying fundamentals remained strong.
He argued that investors who sell based on a single short-term data point [$50B ARR] are "essentially no different from people gambling in a casino." (eg. Hong Kong Economic Journal):
- Commenting on investors' disappointment in OpenAI’s latest revenue forecast: "by itself, not that bad" on their $50B ARR. (Business Times)
And connected investors' disappointment to a failure to understand the business and excessive expectations like a casino.
- Tan rejected the proposition that OpenAI was overvalued. It quotes him saying:
"It’s almost like they are in a casino gambling" in regards to investors' unrealistic expectations and lack of understanding on OpenAI.
- Hock Tan if OpenAI and Anthropic achieve widespread market adoption: their attractiveness would substantially strengthen their bargaining power when raising capital.
But cautioned elevated global interest rates could make capital requirements a constraint for future growth.
- The real driver, he said, is whether frontier AI companies can create models that generate enough value and returns to justify the buildout. (which is fair)
- Most companies are wasting money on tokens. He also warned that most companies just waste money spending tokens like kids in a video arcade.
But Western media commentary narrowed this down to: - "Shortage of capital now, AI Investors are gamblers".
In the conference, Broadcom's CEO was actually defending frontier lab valuations and giving cautiously bullish statements about AI infrastructure demand.
While citing rising interest rates as a potential constraint that larger revenue or adoption could overcome.
Cherry picking quotes out of context disorts the actual messaging.
TLDR:
- Large and elevated AI semi demand, semi supply insufficient to meet market demand. At least another 2 years of elevated spend
- OpenAI valuations defensible, people who sell based on extreme expectations for the company are like casino gamblers.
- Tan stated markets had misinterpreted OpenAI's ARR because its underlying fundamentals remained strong.
- Frontier labs if they achieve widespread adoption / revenue would strengthen capital raising, which overpower headwinds from raising interest rates.
Lot of $AVGO CEO remarks looks heavily nuanced, which Western media coverage distored.
💬 引用 Serenity (@aleabitoreddit)
Wow this is sad to see Korean markets with $SKHY / Samsung crash like this.
Special thanks to $AVGO CEO adding fuel the fire…
Calling AI investors gamblers and saying enterprise AI isn’t delivering ROI. https://t.co/fHzsD9Na4s