The new research paper, written by Yaniv Markovski, formerly of OpenAI, Aviram Roisman, Quack CTO and Co-Founder, and Doron Pryluk, Quack COO, outlines how AI will transform every layer of customer support, success, and experience, ushering in a completely new paradigm for how companies serve their customers by mid-2028
SAN FRANCISCO, Oct. 21, 2025 /PRNewswire/ — Today, Yaniv Markovski, formerly of OpenAI, Aviram Roisman, Quack CTO and Co-Founder, and Doron Pryluk, Quack COO, released a research paper entitled CX 2028. The paper provides an explanation of how customer support will transform step by step in the coming years, with advances in AI powering its shift from deflecting customer support issues to proactively detecting and solving them, reflecting infinite scalability by 2028.
Customer experience is at a turning point. Today, even with the help of AI, scaling customer support remains a major limitation to sustainable business growth. But we’re at the precipice of unlocking AI’s full potential. For customer support, what began as simple knowledge bases and AI-powered scripted replies is transforming into AI integration at every layer of CX. In fact, Gartner predicts that agentic AI will resolve 80% of common customer support issues alone by 2029, reducing operational costs by 30%. Roisman and Pryluk’s new research paper places bets on the development of AI-human agent hybrid teams, the emergence of a reliance on ‘signals’ for proactive detection of CX issues, and additional customer support role shifts, leading to infinite scale in just a few years’ time.
Emerging AI found a natural use case in customer support. From the introduction of ChatGPT until today, AI has been applied primarily in the form of co-pilots that, through their ability to interact directly with users via simple, templated replies, have built confidence and trust in the new technology. As search bars became increasingly irrelevant in favor of AI agents, customer patience for the search process and traditional help centers dwindled. In response, CX teams today are applying AI tools to deflect support tickets and provide direct, instant solutions. But AI applications in customer experience have remained stuck in this automation phase and have not yet shifted towards reinvention.
Roisman and Pryluk predict a future in which AI-led shifts enable resolution, rather than deflection, to become the new customer service norm. They envision a full reimagining of customer support, customer success, and customer experience, along the following timeline:
- By the end of 2025, companies will overcome existing LLM inconsistencies to shift away from strict workflow-, IFTT-, and rule-based automations. This will introduce policy-based approaches, much closer to how human CS agents are trained, enabling the start of a focus on resolving CX issues.
- In early-mid 2026, AI will begin taking on most of the actual workload of support operations for the first time, and AI and human agents will form hybrid teams that serve customers proactively. Personal AI assistants will emerge and AI entities will communicate directly with one another.
- By late 2026, CS will focus on proactively detecting ‘signals’ – indicators like user frustration or backend errors – before customers open tickets. Traditional success metrics like First Time Resolution (FTR) and Time to Resolution (TTR) will be replaced by Proactive Resolution Rate, reflecting the percentage of relevant signals AI agents detected and addressed before they impacted the user experience.
- In early 2027, traditional customer support roles will be replaced by new positions like Customer Expert and Customer Engineer, empowering customer-facing teams to become company growth drivers.
- By early 2028, AI agents will be dogfooding, serving as the first internal users and beta testers of AI-powered customer support. Support agents will find themselves not only helping customer agents and customers’ personal AI assistants, but also internal AI agents.
- In mid-2028, the inflection point will be reached. The concept of support will be entirely transformed as AI-powered teams achieve infinite scalability. The hiring of human experts will be decoupled from customer volume, unlocking the true potential of LLMs’ limitless scalability.
“Customer support is shifting at match pace with AI. We believe that in just a few years, we’ll move well beyond the first sparks of AI-driven deflection to a future where hybrid human-AI teams build, monitor, and even become their own customers, proactively resolving CX issues before customers lift a finger to report them,” said Aviram Roisman, Quack CTO and Co-Founder. “Along the way, job titles, metrics, and the very idea of ‘support’ will be entirely redefined until infinite scalability becomes a reality. Only then will the true potential of customer support to drive business growth be unlocked – and the sky’s the limit.”
Read CX 2028 here. You can also register here for Roisman and Pryluk’s Road to CX 2028 masterclass at Elevate 2025 by monday.com in London on October 23rd or Pryluk’s CX 2028 workshop at the AI for Customer Support Summit in Boston on October 29th here.
About Quack AI
Quack is the only proactive agentic AI platform purpose-built for SaaS companies, transforming customer support from a reactive cost center into a growth engine. Quack enables CX teams to train, deploy, and manage unlimited AI agents that deliver instant, contextual, human-grade support across every channel, tailored to the pace and complexity of software companies. Quack unifies multi-channel interactions, knowledge management, QA, and Voice of Customer insights to provide every AI superpower support teams need within one CRM-agnostic operating system. Founded by Nadav Kemper and Aviram Roisman, Quack maintains a global presence in Tel Aviv, San Francisco, and New York. Visit www.quack.ai or follow us on LinkedIn for more information.
Press Contact
Kate Schoenstadt
Headline Media
kate@headline.media
+972 54 777 6684
SOURCE Quack