高盛：莎莎下半年度毛利率受压 续吁“沽售”

Artificial intelligence (AI) is evolving so rapidly that it's starting to feel like we live in a sci-fi movie. First, there was predictive AI that analyzes data and uses machine learning algorithms to forecast future outcomes. Then, we moved to generative AI that creates new content like text, images, and music. Now, we’ve arrived at the agentic AI stage — not only does AI generate content, but it’s also able to be conversational and autonomously act and react.

What sets autonomous agents apart from its predecessors is that it can reason not only based on predictions it makes from large datasets, but also based on their ability to perceive their environment and then take autonomous action, and even learn from feedback and adapt.

Agentic AI and the AI agents that help it execute tasks are expected to be top strategic technology trends . This evolution emphasizes autonomy and adaptability. Agentic AI is poised to transform industries like healthcare, finance, and manufacturing by seamlessly integrating with data platforms and helping with time-consuming jobs. Imagine AI that can act as digital labor, making decisions and adapting to new situations with remarkable efficiency.

What we’ll cover:

• Agentic AI, defined
• How does agentic AI work?
• Benefits of agentic AI
• Agentic AI and data platforms: a fundamental partnership
• What’s next for agentic AI?

Agentic AI is the technology that powers AI agents so they can act autonomously without human oversight. By serving as a comprehensive platform, agentic AI facilitates seamless interaction between AI agents and humans, fostering a collaborative environment where both can work together. This platform has a suite of tools and services to help AI agents learn, adapt, and collaborate so they can quickly handle complex and dynamic tasks. It’s the next frontier of AI known for its ability to operate independently by making decisions, adapting to dynamic situations, setting goals, and reasoning.

One of the standout features of agentic AI is that it can simplify the development and deployment of AI agents, so integrating AI into daily operations is less resource-intensive. With this framework, businesses can customize AI agents to meet their specific needs, whether those needs involve automating repetitive tasks, enhancing customer service, or driving strategic decision-making.

Where traditional AI systems are rigid and struggle with complex, multistep tasks, agentic AI can be flexible and adaptable. The platform's flexibility and adaptability ensure that AI agents can be tailored to a wide range of industries and applications. Through natural language processing (NLP), agentic AI systems like Agentforce can mimic human behavior, making them ideally suited to handle a wide range of intricate and ever-changing situations.

There are three main features that contribute to agentic AI:

1. Autonomy: Agents can perform tasks on their own, without human oversight or direction.
2. Adaptability: They can learn from their interactions, receive feedback, and change their decisions based on what they've learned.
3. Goal orientation: They can take specific tasks and reason about how to achieve them.

Agentic AI has the potential to change how we interact with technology. Though it’s still early, the groundwork being laid today points to a future where AI agents operate with the autonomy and adaptability of their human counterparts. This transformative potential promises to improve efficiency and bring new opportunities for innovation in almost every aspect of our lives.

Agentic AI operates through a complex network of autonomous software components known as "agents" that draw from massive amounts of data and learn from user behavior to improve over time. Each agent is designed with specific goals and abilities, working in harmony to tackle complex tasks. This innovative approach to agentic AI relies on a blend of advanced technologies such as machine learning, NLP, and knowledge representation that helps agents learn, communicate, and reason effectively.

The underlying architecture of agentic AI sits on a distributed systems platform, which ensures scalability and ?high performance. This platform allows multiple agents (or multi-agent) to operate simultaneously across different servers to enhance the system's overall efficiency and reliability. Agents can collaborate in real-time by sharing information and coordinating their actions to more intuitively and seamlessly interact with users.

Agentic AI uses self-contained agents — AI models that autonomously complete tasks and manage workflows using machine learning, algorithms, and predictive analytics to make real-time decisions.

It follows a five-step process to solve complex problems:

1. Perceive: AI agents gather and decode information from sources like sensors, databases, and interfaces to turn data into insights. They pinpoint meaningful patterns and extract what’s most relevant in their environment.
2. Reason: A large language model (LLM) guides the reasoning process — understanding tasks, crafting solutions, and coordinating specialized models for jobs like content generation or image analysis.
3. Act: Agents perform tasks by connecting with external systems through APIs. Built-in guardrails ensure safety and compliance — like limiting insurance claim processing to specific amounts before human review.
4. Learn: Agents evolve through feedback and get better with every interaction to refine decisions and processes. This continuous improvement drives smarter performance and greater efficiency over time.
5. Collaborate: Agent builder tools are designed to make it easy to deploy agents by lowering the barrier to entry for both experienced programmers and newcomers alike. The open-source nature of the agentic builder tools further encourages a dynamic community of contributors who continuously make the platform better. This collaborative effort not only accelerates the development of new features but also ensures that the technology remains at the forefront of AI innovation.

Overall, the operational mechanics of agentic AI are designed to foster autonomy, adaptability, and efficiency. By using advanced technologies and a collaborative, open-source approach, agentic AI will change various industries and roles, improving the way we interact with technology.

For example, in customer service, autonomous agents can further personalized interactions, offer proactive service, and handle multichannel support. In sales, AI agents can gain leads and move them down the pipeline, book meetings, and answer questions, day or night. Even in marketing, autonomous agents can handle campaigns from creation to optimization. They can build journey maps, and they can also analyze performance metrics and recommend ways to improve a marketing plan.

Agentic AI software is a type of artificial intelligence (AI) that can operate independently, making decisions and performing tasks without human intervention. These systems are able to learn from their interactions and adapt to new situations, improving their performance over time.

Agentic software is particularly useful in environments where tasks are complex, data is abundant, and real-time decision-making is crucial. Examples of AI agent software in action include AI chatbots that can handle customer service inquiries and recommendation systems that personalize content or products.

Agentic AI represents a leap forward in AI that sets it apart from its predecessor AI systems.

One of its standout advantages is that it improves the adaptability and efficiency of AI applications. Agentic AI can significantly speed up development processes by automating tasks and making decisions independently. This autonomy saves time and reduces operational costs. Moreover, the complex algorithms at the heart of agentic AI ensure improved overall performance, making it an asset in any industry.

One of the most compelling advantages of Agentic AI is its potential to create more personalized and engaging interactions for users. By mimicking human-like decision-making, agentic AI can offer intuitive and seamless interactions. Whether it's in customer service, healthcare, or education, the ability to provide personalized customer support can lead to higher user satisfaction and loyalty.

Autonomous AI agents process vast data streams in real time, providing actionable insights for smarter decisions. By detecting patterns and forecasting outcomes, they enable faster, data-driven choices with confidence.

Agentic AI automates repetitive tasks and streamlines workflows, freeing up human teams to focus on strategic, high-value work. With real-time decision-making and continuous learning, these agents complete tasks faster and with greater accuracy, boosting overall efficiency.

At its heart, agentic AI is revolutionizing work across industries and within job functions by using, learning, and building on enterprise knowledge to drive workflow automation. Now, with more advanced capabilities for bringing disparate and diverse data sources together and making it available through a common metadata framework that speaks the same language, businesses can use the value of all their data to automate complex tasks and make real-time, data-driven decisions. This seamless integration improves efficiency and clears the way for new avenues of innovation and growth.

Combining agentic AI with Data Cloud, coupled with retrieval augmented generation (RAG), enables businesses to use both structured and unstructured data to better serve customers.

Data Cloud's advanced data tools, combined with autonomous decision-making, work seamlessly together for smarter outcomes. This collaboration facilitates the creation of intelligent applications to provide a competitive edge by adapting to dynamic business environments.

In addition to the ability to access a wide range of data, this partnership simplifies the deployment process, which improves the overall user experience. As a result, businesses can focus on their core competencies while these advanced AI capabilities drive strategic initiatives. Additionally, the agentic AI layer continuously learns and evolves as the system processes more data. This continuous learning loop means that the AI system can adapt to new data to deliver precise insights and offer more intelligent decision-making in response to evolving conditions and demands.

With the advantage of being able to fully access their data and combine it with AI, businesses in finance, healthcare, retail, and manufacturing can simplify operations and improve customer engagement across sales, service, marketing, and commerce.

Agentic AI is well-positioned to revolutionize the world. Its compatibility with existing systems, ability to create personalized user experiences, and robust security features make it an indispensable tool for the future.

Innovations that will transform industries such as sales, service, marketing, and commerce. Gartner predicts that “by 2028, 15% of day-to-day work decisions will be made autonomously through agentic AI, up from 0% in 2024.”

Imagine a world where AI agents can seamlessly handle complex customer inquiries, adapt marketing strategies in real-time, and optimize supply chains with unparalleled efficiency. The potential for agentic AI to streamline operations and enhance customer experiences is immense.

One of the most exciting aspects of agentic AI is its ability to learn and improve over time. As these AI agents accumulate more data and experience, their decision-making abilities will become increasingly sophisticated. This continuous learning process will help businesses? stay ahead of the curve, responding quickly to market changes and customer needs. The integration of agentic AI with data platforms will be a game-changer, providing seamless access to vast amounts of information and enabling more informed and timely decisions.

Moreover, the impact of agentic AI extends beyond individual industries. It has the potential to reshape entire ecosystems, fostering collaboration and innovation on a global scale. As more organizations adopt this technology, we expect to see the emergence of new business models and partnerships that make the most of the unique capabilities of agentic AI. This collaborative approach will drive exponential growth and create new opportunities for businesses and consumers alike.

Agentforce, as an example of agentic AI, is the agentic layer of the Salesforce Platform that helps companies get more done, enables representatives to build better customer relationships, and serves as an always-on digital workforce for AI success.

The future of agentic AI isn't just about what is possible today, but the endless possibilities that lie ahead.

Sarah Sung is a senior writer at Salesforce. Previously, she was a lifestyle writer covering everything from AI and tech to health and wellness to food and drink for publications including the San Francisco Chronicle and AFAR. She has also worked in content marketing at Under Armour, Gap, and Travelocity.