Generative AI technology has penetrated multiple domains over the last few years. Much of this progress is due to advances in new large-scale language models made possible by Transformers. Meanwhile, improvements to slightly older technologies have made it easier for AI to generate higher quality text, images, voice, synthetic data, and other types of content.
The recent introduction of ChatGPT has put generative AI in the spotlight, raising public awareness of its potential for business, productivity, and art.
“With the release of ChatGPT, anyone with a browser can access AI for free, so families, children, and people with no background in AI or data science can take advantage of ChatGPT. ”said Bret Greenstein, Data and Analytics Partner at PwC. “This follows on from his year of image-generating AI and filters for his mobile app that produces magical results. increase.”
Jonathan Watson, CTO of legal practice platform Clio, also attributes the surge in generative AI to recent advances in generative models, such as generative adversarial networks and variational autoencoders, capable of producing high-quality outputs. . In addition, generative AI has many more attractive applications for more people, such as music, art, games, and healthcare.
Some observers have called generative AI a new general-purpose technology that could have the same kind of wide-ranging impact as steam engines and electricity. Rex Chekal, Principal Product Designer at TXI, said: A product innovation company based in Chicago. “Basically, it frees up my cognitive bandwidth so I can focus on high-impact and high-value tasks.”
Generative AI Industry Use Cases
Beyond composing emails and asking questions faster, industry professionals are seeing many new generative AI use cases.
- Generate code. Partner and chief data scientist Donncha Carroll, who leads the data science and engineering team at Lotis Blue Consulting, said his group uses GitHub Copilot to write entire blocks of Python code to support the service. increase. Depending on the project, using this tool can increase productivity by 30% to 50%.
- Write different types of text. John Blackmon, CTO of ELB Learning, said his company uses generative AI to generate different types of content, including resource guides, how-to articles, news articles, essays, product descriptions and social media posts. says that “As long as you overhaul your AI content and use it as a supplement rather than the final content, it can make your life and work easier,” he said.
- automation. Kavitha Chennupati, senior director of global product management at IT solutions developer SS&C Technologies, said the company is using generative AI to suggest where new automation should be deployed. This will enable more workers to start developing automations such as robotic process automation bots and low-code driven processes.
- documentation. Pierre Custeau, CTO of ToolsGroup, a supply chain planning and optimization company, uses generative AI tools to aid the process of creating better documents.
The future of generative AI
Today, virtually all enterprise apps and services have adopted generative AI to some degree. And while this technology has great potential, businesses must consider several challenges and limitations as they expand their use of the technology. Many of the first limitations slow down your app, but others can cause real problems, such as AI hallucinations, where generative AI apps create content that isn’t tied to fact. Recent examples of AI hallucinations include a Google bard who mistakenly said the James Webb Space Telescope took the first picture of an exoplanet, and an Australian mayor who said ChatGPT was imprisoned for bribery. After that, he sued OpenAI for defamation.
Businesses using this kind of chatbot should be aware that this kind of misinformation can direct customers to do risky repairs, thus damaging their brand. . Successful companies develop measures to mitigate the potential for false alarms and identify ways generative AI can deliver real value to customers and revenue.
Here are 10 trends to watch in the future.
1. Improvisation
Some of the most notable applications of generative AI are art, music, and natural language processing. Clio’s Watson expects this will increase the need to acquire rapid engineering skills to create better content. He expects more companies to improve his UX through prompt-based creation tools. However, IT decision makers need to protect corporate data and information while using these tools. If implemented correctly, it may not appear to work with AI.
2. APIs unlock enterprise use cases
Chat may be all the rage today, but new APIs make it easier to incorporate a variety of generative AI capabilities into your enterprise apps. PwC’s Greenstein said, “ChatGPT is used for many things, from programming software to bedtime children’s stories, and the APIs that make ChatGPT possible are very interesting.” . With these APIs, any application, from mobile apps to enterprise software, can be enhanced with generative AI. Microsoft and Salesforce are already experimenting with new ways to infuse AI into productivity and CRM apps.
3. Rethink business processes
As generative AI improves, more mundane tasks may be automated or augmented. Greenstein predicts that this will enable companies to rethink business processes and use technology to expand what their employees can do. “This will lead to the emergence of entirely new business models, just like after any disruptive technology hits the market,” he said. “AI-native business models and experiences will make small businesses look like big companies and help big companies move faster.”
4. Improved Health App
Chekal of TXI sees the potential for generative AI to improve patient outcomes and make lives easier for healthcare workers. Generative AI extracts and digitizes medical documents, enabling healthcare organizations to access patient data more efficiently. It also improves personalized medicine and treatment by organizing more medical, lifestyle, and genetic information for the right algorithms. Intelligent transcription saves time and helps summarize complex information as part of the doctor-patient conversation rather than as a separate process. It also improves patient engagement through personalized recommendations, medication reminders and better symptom tracking.
5. Better synthetic data
Synthetic data has been around for years. Improvements in generative AI technology could help businesses find ways to take advantage of imperfect data while easing privacy concerns and regulations. “Using generative AI to create synthetic data can greatly improve our ability to rapidly create new AI models, enhance decision-making capabilities, and provide ways for organizations to respond to change in a more agile manner.” “There is a possibility,” he said ToolsGroup’s Cousteau. .
6. More effective scenario planning
Cousteau also believes generative AI could improve our ability to simulate large-scale macroeconomic and geopolitical events. The industry is grappling with a series of events that have caused major disruptions to his chain of supply with long-lasting effects on organizations, economies and the environment. Custeau’s team has been looking for better ways to simulate rare events. This helps mitigate its negative effects in a cost-effective manner.
7. Hybrid model for increased reliability
Large Language Models (LLMs) such as ChatGPT show the potential of new technologies such as Transformers. However, future advancements may often require a combination of multiple models. Emmanuel Walckenaer, CEO of his content generation platform Yseop, said: Hybrid models combine the benefits of LLM with the precise and controllable narratives of symbolic AI. He predicted that hybrid models would drive innovation, productivity and efficiency within regulated industries by ensuring more accurate outputs.
8. Personalized generative applications
ELB Learning’s Blackmon predicted an increase in generative applications customized to individual user preferences and behavioral patterns. For example, a personalized generative music application may create music based on a user’s listening history and mood. Similarly, AI can analyze an individual learner’s strengths, weaknesses and learning styles during her online training and recommend the most effective teaching methods and the most relevant resources. Ultimately, AI-powered virtual assistants could become a standard feature of learning platforms by providing real-time support and feedback as learners progress through courses. Personalized assistants in enterprise apps can help streamline work processes based on your personal style.
9. Domain-specific apps
Carroll of Lotis Blue Consulting believes generative AI will open up many opportunities for fine-tuning domain-specific applications. For example, generative AI can extract insights from medical publications about medical conditions or automate tedious query-response entry tasks in customer service centers. LLMs can capture industry-specific information and provide insight into domain-specific workflows. For an IT decision maker, the focus is from exploring cool new technologies to providing great data to train customers on his LLM of applications without introducing operational or reputational risk into the process. Moving on to identify. “This could be the catalyst IT leaders need to change the data quality paradigm and make their case for business investing in building high-quality data assets,” he said. I’m here.
10. Natural Language Interface
Todd Johnson, managing director of digital transformation consultancy Nexer Group, predicts that generative AI will help drive the creation of more intuitive and easy-to-use natural language interfaces (NLIs). . “NLI allows users to communicate with computer systems using natural language instead of programming languages and syntax,” he explained. For example, in the context of his chain of supply, generative AI could provide an audio interface for warehouse distribution center workers. A worker can interact with her NLI via a headset connected to the manufacturer’s ERP system, navigate a packed warehouse, locate specific items, and reorder materials and supplies. I made it. This reduces clerical errors and increases efficiency.
“You will hear the word. co-pilot “This technology empowers everyone to focus on how to better serve their customers and grow their business,” Johnson said.
