Netflix Founder Goes to His Most Feared Place

marsbit發佈於 2026-04-20更新於 2026-04-20

文章摘要

Netflix co-founder Reed Hastings is stepping down as chairman after nearly 30 years with the company, despite Netflix reporting its strongest-ever quarterly results with a 16% revenue increase and 83% surge in net profit. Hastings, who studied AI in the 1980s, has joined the board and donated $50 million to study AI’s impact on society. He now views AI as the biggest threat to Netflix, fearing that AI-generated content could eventually make free platforms like YouTube more appealing than paid services. While Netflix is using AI to improve production efficiency, Hastings is hedging his bets by engaging directly with AI governance, concerned that the rapid pace of AI development could disrupt the entire content industry.

Author: David

Netflix has never been as profitable as it is now, yet its founder chose this moment to leave.

On April 16, Netflix released its Q1 2026 earnings report, with revenue of $12.25 billion, up 16% year-over-year, and net profit surging 83% YoY. Earnings per share were $1.23, nearly 60% higher than Wall Street's expectation of $0.76.

But the earnings report also announced another thing: co-founder and current Chairman Reed Hastings will not seek re-election after his term ends in June.

Hastings founded Netflix in 1997, building it from a DVD-by-mail business into a streaming giant with over 325 million paid members worldwide, working for nearly 30 years. In 2023, he handed the CEO role to his successor and stepped back to Chairman. Now, he's leaving the Chairman role too.

In its filing with the U.S. Securities and Exchange Commission, Netflix specifically wrote: "This decision is not the result of any disagreement with the company."

But the more they emphasize no disagreement, the more it makes people wonder what he is actually going to do.

A little-known fact is that in May last year, Hastings had already joined the board of Anthropic. For almost 30 years, his business has essentially been about getting people to pay for content, while Anthropic's Claude, though not directly generating video, is changing the way content is produced.

From text to images to video, costs are getting lower and speeds faster.

Netflix's profitability relies on good content being worth paying for. If AI lowers the barrier to content creation enough, does that premise still hold?

Hastings is clearly already thinking about this.

What is he afraid of?

As a top global content producer and distributor, Netflix's founder has always had a conceptual concern about AI.

You might not know this, but in 1988, Hastings was studying for a master's in AI at Stanford. That's right, 40 years ago he was researching artificial intelligence. It's just that the AI of that era was nothing like the useful tool it is today...

In 2022, Hastings was invited as a speaker at Stanford's graduation ceremony.

He later talked about it himself, sounding like he was telling a joke about a wrong turn in his youth. AI didn't work out, so he turned to starting a software company, and later founded Netflix, which he ran for nearly 30 years.

Someone who studied AI couldn't help but pay attention to the field.

In a 2024 interview, talking about AI, he was quite relaxed: "AI will help us become more creative, we can use these tools to make more shows." His attitude then was one of embrace. AI was a tool, there to help, not to take jobs.

In March 2025, he donated $50 million to his alma mater, Bowdoin College.

This liberal arts college in Maine doesn't work on large models; Hastings gave them money for a research initiative called "AI and Humanity," specifically studying the impact of AI on work, education, and human relationships.

On the day of the donation, he said something completely different from his relaxed tone a year earlier: "We are going to fight for the survival and flourishing of humanity."

In one year, AI progressed rapidly, and his stance shifted from AI helping work to AI being a threat to humanity.

Two months later, he joined the Anthropic board.

He was appointed by an independent body called the "Long-Term Benefit Trust," whose five members hold no Anthropic stock, with the sole duty of ensuring AI development aligns with humanity's long-term interests.

In March of this year, he spelled it out most clearly in another interview. The host asked him what the biggest risk facing Netflix was, and he skipped over competitors and subscriber growth, saying just two words:

AI.

He said if AI makes the free content on YouTube cool and engaging enough, and all the young people go watch free content, then who will pay for Netflix?

From public information, you can find Hastings calling himself an "extreme techno-optimist." He doesn't think AI itself is bad, the problem is the speed gap.

AI technology is moving too fast, and humanity's moral and institutional systems can't keep up.

This explains his seemingly contradictory choices over the past year. Donating money not to a technical AI lab, but to a humanities college; joining a board, not any commercial AI company's advisory group, but Anthropic's safety committee.

In the author's opinion, Hastings is more qualified than most to be concerned about whether AI will disrupt the industry.

Netflix itself was the disruptor in the last cycle. It used streaming to kill DVD rentals, crippled cable TV, and forced all of Hollywood to rebuild its distribution system. He personally did the "use new technology to drive content and distribution costs low enough to kill the previous winners" thing once.

Now he's looking at AI, probably wondering who's next.

So, Hastings is simultaneously a major shareholder of Netflix and a director at Anthropic. Holding shares of the company he founded, he takes a seat in the industry that might disrupt it.

This might not be called retirement. It's called hedging.

Despite the AI impact, Netflix has actually never been better

Four years ago, Netflix was a company with just over $30 billion in annual revenue and less than 20% profit margins, being hounded by Wall Street asking "when will you make real money?" This earnings report four years later provides the answer.

Q1 2026, net profit was $5.28 billion, up 83% YoY. Free cash flow was $5.09 billion, almost double that of the same period last year. Meanwhile, the profit margin reached 32%. The full-year revenue guidance is $50.7 to $51.7 billion. If they actually achieve it by year-end, it means Netflix's revenue has nearly doubled in three years.

Beyond daily operations, Netflix isn't blind to AI either.

A few weeks ago, it spent up to $600 million to acquire InterPositive, a company that makes AI-assisted film and TV production tools, using AI to accelerate script development, scene previews, and post-production. Netflix also specifically mentioned generative AI in its earnings letter, saying it will use it to improve content production and user experience.

Using AI to reduce production costs and improve efficiency is a sound approach. In fact, the entire Hollywood or content production industry is moving in this direction.

It's just that the concern founder Hastings expressed in the interview might be about a different problem.

In February of this year, ByteDance released its video generation model Seedance 2.0. Upload a photo, and it generates a 2K video with camera movement, sound effects, and lip-syncing in 60 seconds.

After testing it, Feng Ji, producer of Black Myth: Wukong, said four words: "The childhood era of AIGC is over." Director Jia Zhangke posted on Weibo saying he was preparing to use it to make a short film...

More concrete numbers come from within the industry. According to Securities Times reports, in the e-commerce advertising sector, one person using Seedance 2.0 can complete in 30 minutes what used to take 7 people 3 days, reducing costs by over 99%.

Extras in Hengdian, video editors, special effects producers—people across the entire industry chain are all saying the same thing—job anxiety.

Gong Yu, founder of iQiyi, publicly stated a judgment late last year: AI could reduce the costs of the film and television industry by an order of magnitude, increase creators by an order of magnitude, and increase works by two orders of magnitude.

Netflix using AI to reduce production costs is about improving efficiency within the existing model. But what Seedance and others are doing is lowering the barrier to "making video" from millions of dollars to a few dollars.

The future Hastings spoke of, where "free content on YouTube becomes good enough," is step by step becoming reality.

Of course, all this might not be directly related to his decision to leave Netflix now. He started handing over power in 2023—CEO, Chairman, step by step—with at least a three-year transition period.

It's just that the timing is indeed微妙 (subtle/interesting). Netflix delivered its best-ever earnings report, and the stock fell 8% after hours. On the same day, the founder announced his complete departure.

After June, Hastings' name will disappear from Netflix's board list.

His current titles are Director at Anthropic, Director at Bloomberg, and owner of a ski resort in Utah. He still holds Netflix stock; Forbes estimates his net worth at $5.8 billion, mostly tied to Netflix.

He holds Netflix's money, sitting at AI's table.

As for whether this choice is foresight or overcaution, we might only get the answer when AI can actually produce a movie that audiences are willing to watch to the end.

相關問答

QWhy did Reed Hastings decide to step down as Chairman of Netflix despite the company's strong financial performance?

AReed Hastings stepped down as Chairman of Netflix to focus on his role at Anthropic's board and to address the long-term risks and opportunities posed by AI, which he sees as a potential threat to the traditional content business model.

QWhat is Reed Hastings' background in AI, and how does it influence his current concerns?

AReed Hastings studied AI for his master's degree at Stanford in 1988. His early exposure to AI fuels his deep concern about its rapid advancement and potential to disrupt industries, including content creation and distribution.

QHow does AI pose a threat to Netflix's business model according to Hastings?

AHastings fears that AI could make free content on platforms like YouTube so compelling and high-quality that consumers may stop paying for Netflix, undermining its subscription-based revenue model.

QWhat actions has Hastings taken to address the challenges and opportunities of AI?

AHastings joined the board of Anthropic, a company focused on AI safety, and donated $50 million to Bowdoin College for a research program studying AI's impact on work, education, and human relationships.

QHow is Netflix itself responding to the rise of AI in content creation?

ANetflix acquired InterPositive, an AI-assisted film production tool company, for up to $600 million and is using generative AI to improve content production efficiency and user experience, aiming to reduce costs and enhance creativity.

你可能也喜歡

如何让自己变得让人工智能永远也无法取代

面对人工智能的冲击,许多人担心工作被取代。然而,真正的威胁在于个人对他人和系统的依赖,以及由此产生的“薪资奴役”——即为生存而从事无意义、枯燥的工作。摆脱这种困境的关键,不是抵制技术,而是成为拥有高自主性的“不可受雇”个体。 文章提出了成功抵御AI替代的五个核心要素:自主性(主动行动的能力)、品味(判断事物价值的经验)、说服力(让他人关注你工作的能力)、毅力(坚持并从错误中学习)和迭代(根据反馈持续改进)。这些能力无法仅通过理论学习获得,必须通过实践来培养。 要启动转变,首先要彻底改变环境,重塑身份认同。其次,应选择一个能获得真实、快速反馈的实践领域,例如创业。在众多技能中,内容创作(媒体)比编写代码更具优势,因为其价值是主观的,需要独特的审美和判断力,这正是AI目前难以完全复制的。 具体行动上,可以从三个步骤开始: 1. **挖掘原始素材**:反思自己长期痴迷的知识领域、轻松解决的难题或童年被压抑的兴趣,找到独特的个人经验。 2. **确立反向思考主轴**:找出你坚信但主流观点错误的地方,或行业内普遍忽视的“皇帝新衣”,形成独特的批判性视角。 3. **立即发布**:将前两步的思考融合,撰写并发布第一个核心内容(如帖子、视频),勇敢接受真实世界的反馈,并在此基础上持续学习和迭代。 最终,抵御AI的关键在于构建一份与自身身份深度契合的毕生事业,通过持续的内容创作和真实互动,建立无法被自动化取代的独特价值和影响力。行动,从今天发布第一个想法开始。

marsbit1 小時前

如何让自己变得让人工智能永远也无法取代

marsbit1 小時前

通过掷骰子离线保管比特币密钥:并非人人愿意为之

文章探讨了通过投掷骰子生成比特币钱包种子短语的安全方法及其现实挑战。核心观点如下: **1. 骰子提供物理熵源** 骰子结果由众多微小变量决定,理论上虽可预测,但实践中无法被攻击者复制或计算,从而提供高质量的随机性。每个六面骰子投掷约产生2.585比特熵,50次投掷即可满足典型12词助记词(128比特熵)的安全需求。 **2. Coldcard漏洞事件凸显手工熵源的价值** 近期Coldcard硬件钱包因固件漏洞导致其内部随机数生成器存在缺陷,致使约1128枚比特币被盗。但那些**完全**通过足量骰子投掷生成种子短语的用户未受此漏洞影响,因为他们的主密钥未使用有缺陷的生成器。 **3. 重要警示:手工种子并非万能保护** 安全研究员指出,即使用户使用骰子生成了安全的种子,若他们使用了Coldcard的其他功能(如生成纸钱包、克隆密钥、共享签名密钥、密码等),这些**衍生密钥**仍可能调用有漏洞的随机数生成器,从而存在风险。安全种子不保证设备生成的所有秘密都安全。 **4. 手工生成熵源的现实局限性** 尽管数学上可靠,但该方法对大多数用户并不友好: * **过程繁琐易错**:需投掷50-99次,精确记录,任何输入错误都会导致钱包完全不同。 * **引入新风险**:用户可能在记录、转换过程中泄露信息,或使用有偏的骰子/投掷方式。 * **用户体验差**:难以想象大规模推广需要用户手动投掷近百次骰子。安全措施需适应现实生活场景和普通用户的知识水平。 **5. 给用户的建议** 受影响的Coldcard用户应: * 更新固件至最新版。 * 检查是否使用过有漏洞的功能生成了次级密钥或密码,如有则需立即更换。 * 考虑采用多签方案,使用不同厂商的设备分散风险。 **结论**:手工投掷骰子生成熵源是技术娴熟用户的一个有效安全选项,但其过程复杂、容易出错,不适合作为主流用户的默认方法。长远目标是依赖安全、透明且无需专业知识的硬件/软件随机数生成方案。

cryptonews.ru4 小時前

通过掷骰子离线保管比特币密钥:并非人人愿意为之

cryptonews.ru4 小時前

交易

現貨
活动图片