Google CEO Sundar Pichai arrives to attend the bipartisan US Senate Artificial Intelligence (AI) Insights Forum at the US Capitol on September 13, 2023 in Washington, DC.
Mandel Gunn | AFP | Getty Images
“We’re moving from talking about AI to applying AI at scale,” Microsoft CEO Satya Nadella said on Tuesday’s earnings call. “By bringing AI to every layer of our technology stack, we are winning new customers and driving new benefits and productivity gains.”
Last year marked the beginning of a generative AI boom, with companies racing to incorporate increasingly sophisticated chatbots and assistants into their core products. Nvidia was making a lot of money. Its graphics processing units (GPUs) are at the heart of large-scale language models created by OpenAI, Alphabet, Meta, and a growing number of heavily funded startups vying for a piece of the generative AI pie. Masu.
As 2024 approaches, executives are outlining plans for continued investment in AI and providing greater clarity to investors about their company’s strategy. Based on the latest financial results, one of the key priority areas is AI-as-a-service models, or large-scale AI models that clients can use and customize according to their needs. Another is investing in AI “agents.” The term is often used to refer to tools ranging from chatbots to coding assistants and other productivity tools.
Overall, executives were reminded that AI is no longer just a toy or a lab concept. It’s really here.
For the largest enterprises, there are two major areas of investment: AI efforts and the cloud infrastructure needed to support large-scale workloads. To get there, cost cutting will continue in other areas as well. It’s a message that has become familiar over the past few quarters.
Meta CEO Mark Zuckerberg on Thursday highlighted the company’s continued AI efforts alongside broader cost reductions.
Mark Zuckerberg, founder and CEO of Meta, speaks at the MetaConnect event at Meta Headquarters in Menlo Park, California, on September 27, 2023.
Josh Edelson | AFP | Getty Images
“In 2023, we are focused on improving our business to make Meta a stronger technology company and provide the stability to realize our ambitious long-term vision for AI and the Metaverse,” Zuckerberg said in an earnings call. It was the “year of efficiency,” he said.
Nadella told investors that Microsoft will expand its AI investments and cloud efforts, even if it means carefully considering expenses in other departments, with “disciplined cost management across all teams.” He said he is working hard to.
Microsoft CFO Amy Hood emphasized the company’s “consistency in re-converting our workforce to the AI-first work we are doing without making significant additions to our workforce,” and said the company plans to move forward. “We will prioritize investment in AI as a future issue.” To shape the next decade. ”
The theme is similar at Alphabet, where Sundar Pichai talked about the company’s “focus and discipline” in prioritizing scaling up AI in search, YouTube, Google Cloud, and more. He said investments in infrastructure such as data centers were “key to realizing our big AI ambitions,” adding that the company had cut back on low-priority projects and invested in automating certain processes. .
“We will continue to invest responsibly in data centers and compute to support a new wave of growth in AI-powered services for us and our customers,” Pichai said. “You’ve heard me talk about our efforts to permanently redesign our cost base and improve speed and efficiency. Those efforts continue.”
Within Google Cloud, Pichai said he will reduce expenses by reallocating resources to the most important projects, slowing the pace of hiring, improving technology infrastructure and leveraging AI to streamline processes across Alphabet. . He said total capital spending in the fourth quarter was $11 billion, primarily due to investments in infrastructure, servers and data centers.
Ruth Porat, Alphabet’s head of finance, said the company will continue to invest heavily in “long-term opportunities” with AI and AI applications within DeepMind, with full-year capital spending in 2024 “lower than 2023.” He reiterated that he expects the economy to grow significantly.” Provide cloud and other systems.
Amazon CEO Andy Jassy said on an earnings call this week that generative AI “will ultimately drive tens of billions of dollars in Amazon’s revenue over the next few years.”
As Amazon pours more money into LLM, other generative AI projects, and the necessary infrastructure, AI will continue to be a key investment area for the company, driving increased capital spending this year. Jassy highlighted Amazon’s AI chip efforts, citing customers such as Anthropic, Airbnb, Hugging Face, Qualtrics, and Snap.
Apple CEO Tim Cook has cited generative AI as a key investment area for his company, hinting at an announcement later this year.
“As we look to the future, we will continue to invest in these and other technologies that will shape the future,” Cook said on a call with analysts. “This also includes artificial intelligence, where we continue to invest a significant amount of time and effort, and we look forward to sharing more details about our work in this area later this year. ”
Cook added: “Without going into too much detail, let me just say that we think there’s a huge opportunity for Apple in Gen AI and AI.”
Investors want to know how key infrastructure companies are investing in AI, but they also want to know where and how their money is being made.
Jassy said enterprise customers are looking to use existing models that they can personalize and build on, citing Amazon’s Bedrock as a key focus.
“What we’re seeing is that customers want choice,” Jassy said. “They don’t want just one model to rule the world. They want different models for different applications. And they want different sized models because they have different cost structures and different latency characteristics. We want to try all the models.”
Andy Jassy on stage at the 2022 New York Times Dealbook on November 30, 2022 in New York City.
Toth Robinson | Getty Images
Nadella pointed to Microsoft Azure as a leading “model-as-a-service” product, where customers don’t have to manage the underlying infrastructure and have access to language models large and small, such as Cohere, Meta, and Mistral. We emphasized what we can do. There are also open source options. Nadella said a third of his 53,000 Azure AI customers have joined within the past 12 months.
Alphabet executives highlighted Vertex AI, a Google product that provides more than 130 generative AI models that can be used by developers and enterprise clients such as Samsung and Shutterstock.
Chatter was not limited to LLMs or chatbots. Many technology executives talked about the importance of AI agents, or AI-powered productivity tools, to complete tasks.
Ultimately, AI agents can scan everyone’s calendars to ensure there are no conflicts to schedule group hangouts, book trips and activities, and buy gifts for loved ones. This could be in the form of performing specific job duties, such as outbound sales. However, for now, the tool is primarily limited to tasks such as summarizing, generating to-do lists, and helping you write code.
Nadella is bullish on AI agents, citing Microsoft’s Copilot assistant as an example of an AI application that has “evolved” in terms of productivity benefits and successful business models.
“People will start to think of these tools as something that increases productivity,” Nadella said. “I think this is a new vector for us in what I would call the next phase of knowledge work and frontline work, both in terms of productivity and how we participate.”
Just before Amazon’s revenue took a hit, the company launched Rufus, a generative AI-powered shopping assistant trained on its product catalog, customer reviews, user Q&A pages, and the broader web. Announced.
“The question is, how do we think about generative AI in the consumer business? We’re building dozens of generative AI applications across the company,” Jassy said on the conference call. “Every business we run has multiple generative AI applications being built, and they’re all at different stages, with many launched and others in development.”
Zuckerberg said on the conference call that Meta will also focus in part on building useful AI agents.
“Going forward, our primary goal will be to build the most popular, cutting-edge AI products and services,” Zuckerberg said. “And if we are successful, everyone who uses our services will have a world-class AI assistant to help them get things done.”
Alphabet executives touted Google’s Duet AI, a “packaged AI agent” for Google Workspace and Google Cloud designed to increase productivity and complete simple tasks. Within Google Cloud, Duet AI is helping software developers at companies like Wayfair and GE, as well as cybersecurity analysts at Spotify and Pfizer, Pichai said. He added that Duet AI will soon incorporate Gemini, his LLM from Alphabet, which powers the company’s Bard chatbot.
Pichai said he hopes to provide AI agents that can complete more tasks on behalf of users, including within Google searches, but “there’s a lot more to come.”
“We’re going to use generative AI there as well, especially with cutting-edge models and Bard,” Pichai said. This could allow them to “think forward-looking and act more like an agent over time and go beyond answering and following through with users.”
clock: Megacaps are returning capital.
