ChatGPT, an artificial intelligence-powered chatbot, is having a viral moment, offering a glimpse into the capabilities of a mature technology. It also highlights investment themes in other club holding companies such as NVIDIA (NVDA) and Microsoft (MSFT). At the same time, some people wonder if Google Search will one day come under threat. What is ChatGPT? I asked ChatGPT for clarification. ChatGPT is a variant of the GPT-3 language model specifically designed for conversational text. It is a large-scale language model trained on large amounts of conversational text data, capable of producing more natural and human-like responses in conversation. Like other his GPT models, ChatGPT can generate different styles and formats of text, including answers to questions, explanations, and entire conversations. Of course, that answer is a bit technical and uses acronyms like GPT-3 (short for Generative Pretrained Transformer 3). So I asked ChatGPT to explain it in “easy terms” as well. The answer is ChatGPT, a large-scale language model trained to generate human-like responses to text input. Machine learning algorithms can be used to process and analyze large amounts of conversational text data to generate more natural and human-like responses in conversations. This enables applications such as chatbots to generate responses to questions and statements that are more realistic and engaging for users. ChatGPT was developed by OpenAI, a research startup founded in 2015 where he received a $1 billion investment from Microsoft. ChatGPT launched in his late November and quickly gained attention as people posted screenshots of their responses on his Twitter and other social media platforms. Over 1 million users have signed up for ChatGPT. People asked all sorts of questions, such as whether ChatGPT can explain certain topics and write lyrics. I even saw one of her Wall Street analysts include in a note to a client her ChatGPT-generated poem containing his three buzzwords: semiconductor, research and investor. I have. The first stanza is: ChatGPT is impressive and frankly fun, but the model is far from perfect. OpenAI has an image generation model, DALL-E, but it only generates text and, according to ChatGPT’s own response, doesn’t have the “ability to browse the internet”, so it can’t really answer questions about current events. you can’t. Another limitation at the moment is scale. ChatGPT may return a “Capacity Reached” message when it cannot handle new inquiries due to “extremely high demand”. As with artificial intelligence, prejudice raises ethical concerns. Some people find ChatGPT’s responses sound human-like and confusing, raising questions about its application in the real world and the accuracy of the information it sends. The virality of ChatGPT reveals just how powerful artificial intelligence is and how important it is to the properties of clubs like Microsoft and Nvidia. Microsoft’s involvement Not only has Microsoft invested his $1 billion in OpenAI, but the tech giant’s cloud computing arm, Azure, is the digital land where ChatGPT lives. Microsoft worked with OpenAI to build a startup-only supercomputer that it uses to train models. The supercomputer, announced in May 2020, is hosted in Azure. ChatGPT may be a fun way to generate poetry, but customers appreciate the tools and integrations offered by cloud service providers, so the collaboration between Microsoft and OpenAI has business implications. Microsoft recently incorporated OpenAI’s DALL-E image generation tool into the Azure OpenAI service. Also, Microsoft CEO Satya Nadella said toy maker Mattel (MAT) is already taking his case for commercial use. “Our partnership with OpenAI is a very important partnership for us. OpenAI CEO Sam Altman also praises Microsoft’s role in the startup. Artificial intelligence is playing an increasingly important role in cloud computing, with enterprises using Azure’s AI and machine learning tools for a variety of essential tasks, not just advanced language models. They help companies streamline their operations, generate software code, and build predictive demand models.For example, PepsiCo (PEP) uses Azure Machine Learning (ML) to improve product sales. Machine learning is a subgroup of AI.”Overall, Azure ML revenue has increased by more than 100% for four consecutive quarters,” Nadella said during Microsoft’s Q1 conference call. says. Overall Azure revenue growth has moderated in recent quarters as concerns about a slowing economy permeate among enterprise customers. However, we remain bullish on Azure’s long-term trajectory and believe its artificial intelligence and machine learning tools will help further accelerate cloud adoption as customers harness the potential to improve their own enterprises. I’m here. ChatGPT is just the tip of the iceberg. In that sense, good things for Microsoft could lead to success for other club names, especially chipmakers like his Nvidia and Advanced Micro Devices (AMD). Nvidia, other chip makers and semiconductors are, as we have said before, brick and mortar stores in the digital world, and one of the things that this ChatGPT launch has made clear is that artificial intelligence applications will fill the demand for computing power. It’s about promoting in the years to come. Admittedly, it’s still very early days. With several more years of research and a focus on industry-specific tasks, it is very likely that natural language models will be widely adopted. Automated chatbots and voice systems can provide some assistance, but they are no match for human representatives. ChatGPT seems to be far superior to any chatbot currently in use. We believe that additional research will help bridge the gap between the usefulness of automated and human customer service systems. As these models become more sophisticated and more widely adopted, they require more computing power. This also positions other data center-oriented chip makers such as AMD to benefit. I have no doubt that all these companies will benefit. However, his Nvidia in particular has been a major beneficiary, thanks not only to cutting-edge graphics chip products, but also decades of investment in AI software programming, such as his Omniverse Avatar for intelligent virtual assistants. It will be a bigger driver for adoption. In fact, Nvidia employs more software engineers than hardware engineers, developing software programs that complement high-performance graphics processing units (GPUs) in areas such as AI. Google search threat? Due to ChatGPT’s superior features, some people wonder if ChatGPT could threaten Google search as the dominant online search engine. It’s only natural that it resonates with people. Ultimately, however, we do not believe that ChatGPT poses a real risk to Google’s parent company Alphabet (GOOGL)’s primary source of revenue. This view is shared by analysts at Morgan Stanley. “While we do not deny disruptive technology, we continue to believe that technology must provide solutions that are ten times better than the next 100 million) people,” the analyst wrote. Additionally, the Club, which owns Alphabet, is no stranger to artificial intelligence (more specifically machine learning). Both already play an important role in shaping Google search performance today. Tech giants are investing in more experimental, high-level applications. Alphabet also applies his AI research to many other areas, including cloud services with Google Cloud units, autonomous driving, and more. Morgan Stanley estimates that “about 50% of GOOGL’s workforce is engineers, and industry conversations continue to point to GOOGL as a large corporate technology leader in AI/ML research and investment.” I am,” he wrote. As of September 30, Alphabet had approximately 187,000 employees, according to its quarterly report. Alphabet is working on multilingual models like ChatGPT. In particular, there is Sparrow, his AI assistant in development by DeepMind, which in 2014 was acquired by Google for $600 million. Google also has a conversational language model called LaMDA. First announced in May 2021, followed by a second generation version earlier this year. All three of his models, ChatGPT, Sparrow and LaMDA, have differences but share a basic network architecture called Transformer. Some argue that LaMDA is actually stronger than ChatGPT. However, Alphabet is not open to the public. Recently, an Alphabet employee asked management whether ChatGPT’s virality was a “lost opportunity” for the company, he reported to CNBC. In response to this question, management expressed concern over the damage to the company’s reputation if LaMDA were made public. This is also an important point for investors to consider when considering the threat ChatGPT poses to Google Search. As a shareholder, I don’t want Alphabet to expose its flawed AI model to the public and jeopardize the trust of its billions of users. Conclusion ChatGPT’s state-of-the-art features show the impressive work that club holdings such as Microsoft and Nvidia are doing in the important area of artificial intelligence. Keep in mind: It’s still very early innings for the AI. In his recent Citi memo, he estimated adoption of AI-driven workloads to be only about 10%. This means the runway for growth over the next few years is substantial. In some ways, it’s both a short-term and a long-term story. Adoption is on the rise, showing up in quarterly earnings reports of companies such as Microsoft. However, there are many possibilities along the way. When that happens, it creates more data and increases demand for the goods those club holdings offer. I have some questions about what ChatGPT means for Alphabet, but I’m not worried about Google Search losing its throne anytime soon. If anything, ChatGPT demonstrates the potential utility that conversational AI can have in a clear and viral way. That’s good news for the models Alphabet has under its roof. (Jim Cramer’s Charitable Trust is long on his NVDA, MSFT, AMD, and his GOOGL. See here for the full list of stocks.) Jim Cramer’s CNBC Investing Club subscribers: Receive trade alerts before Jim makes a trade. Jim waits 45 minutes after sending a trade alert before buying or selling shares in his charitable trust portfolio. If Jim talks about his stock on his CNBC TV, he will wait 72 hours after issuing a trade alert before executing the trade. The investment club information above is subject to our Terms of Use and Privacy Policy, along with our disclaimer. No fiduciary duty or obligation exists or is created by your receipt of any information provided in connection with The Investment Club. No specific results or benefits are guaranteed.
Satya Nadella, CEO of Microsoft Corp., speaking at the Microsoft Developers Build Conference in Seattle, Washington, USA, Monday, May 7, 2018.
Grant Hinsley | Bloomberg | Bloomberg | Getty Images
ChatGPT, an artificial intelligence-powered chatbot, is having a viral moment, offering a glimpse into the capabilities of a mature technology. It also underscores our investment thesis. NVIDIA (NVDA) and other club holdings microsoft (MSFT) is an integral partner of the company behind this effort. At the same time, some people wonder if Google Search will one day come under threat.