At SC25, WEKA—best known for bringing high-performance data architectures to AI infrastructure—announced something that feels less like an upgrade and more like a pressure-relief valve for the entire AI industry. The company has taken its Augmented Memory Grid technology from concept to full commercial availability on NeuralMesh. And the timing could not be more relevant.

AI builders everywhere are running into the same wall: GPU memory. It’s fast, it’s precious, and it’s nowhere near large enough for the sprawling long-context models and agentic AI workflows that now dominate the market. The industry has thrown compute, distributed clusters, and clever caching at the problem—yet the wall remains.

WEKA’s answer: eliminate the wall entirely.

Validated on Oracle Cloud Infrastructure (OCI) and other major AI clouds, Augmented Memory Grid expands the available GPU memory footprint by 1000x, turning gigabytes into petabytes, while cutting time-to-first-token by up to 20x. Long-context inference, reasoning agents, research copilots, and multi-turn systems suddenly behave like they’ve been freed from a decade-old hardware ceiling.

It’s not an incremental improvement—it’s a structural rewrite of how AI memory can work.

The AI Memory Wall: Why GPU HBM Can’t Keep Up

The bottleneck isn’t theoretical. High-bandwidth memory (HBM) on GPUs is blisteringly fast but extremely small. System DRAM offers more space but only a fraction of the bandwidth. Once both tiers fill, inference workloads begin dumping their key-value cache (KV cache), forcing GPUs to recompute previously processed tokens.

That recomputation is the silent killer: it burns GPU cycles, slows inference speeds, drives up power consumption, and breaks the economics of long-context AI.

As large language models move toward 100K-token, 1M-token, and agentic, continuously-running interactions, the HBM-DRAM hierarchy collapses under its own constraints. And so far, no amount of clever software trickery has truly solved it.

WEKA’s approach: change the architecture.

Augmented Memory Grid: A New Memory Layer for AI

Instead of forcing GPUs to live inside the rigid boundaries of HBM, Augmented Memory Grid creates a high-speed bridge between GPU memory and flash-based storage. It continuously streams KV cache to and from WEKA’s “token warehouse,” a storage layer built for memory-speed access.

The important detail:
It behaves like memory, not storage.

Using RDMA and NVIDIA Magnum IO GPUDirect Storage, WEKA maintains near-HBM performance while letting models access petabytes of extended memory.

The result is that LLMs and reasoning agents can keep enormous context windows alive—no recomputation, no token wastage, and no cost explosions.

“We’re bringing a proven solution validated with OCI and other leading platforms,” said WEKA CEO and co-founder Liran Zvibel. “Scaling agentic AI isn’t just compute—it’s about smashing the memory wall with smarter data paths. Augmented Memory Grid lets customers run more tokens per GPU, support more users, and enable entirely new service models.”

This isn’t “HBM someday.” It’s HBM-scale capacity today.

OCI Validation: The Numbers That Matter

The technology didn’t just run in a lab. OCI testing confirmed the kind of performance that turns heads:

• 1000x KV cache expansion with near-memory speeds

• 20x faster time-to-first-token when processing 128K tokens

• 7.5M read IOPs and 1M write IOPs across an eight-node cluster

These aren’t modest deltas—they fundamentally change how inference clusters scale.

Nathan Thomas, VP of Multicloud at OCI, put it bluntly:
“The 20x improvement in time-to-first-token isn’t just performance—it changes the cost structure of running AI at scale.”

Cloud GPU economics have become one of the industry’s greatest pain points. Reducing idle cycles, avoiding prefill recomputations, and achieving consistent cache hits directly translate into higher tenant density and lower dollar-per-token costs.

For model providers deploying long-context systems, this is the difference between a business model that breaks even and one that thrives.

Why Long-Context AI Needed This Yesterday

As LLMs evolve from text generators into autonomous problem-solvers, the context window becomes the brain’s working memory. Coding copilots, research assistants, enterprise knowledge engines, and agentic workflows depend on holding vast amounts of information active simultaneously.

Until now, supporting those windows meant trading off between:

• astronomical compute bills

• degraded performance

• artificially short interactions

• forced summarization that loses fidelity

With Augmented Memory Grid, the trade-offs shrink dramatically. AI agents can maintain state, continuity, and long-running memory without burning GPU cycles on re-prefill phases.

Put differently:
LLMs get to think bigger, remember longer, and respond faster—without crushing infrastructure budgets.

A Broader Shift: AI Architecture Moves Beyond Compute

For the last five years, AI scaling strategies have focused overwhelmingly on compute—bigger GPUs, faster interconnects, more parallelization. Memory, by contrast, has been the quiet constraint no one could fix.

WEKA’s move highlights a turning point:
AI’s next leap forward won’t come from more FLOPs. It will come from smarter memory architectures.

NVIDIA’s ecosystem support—Magnum IO GPUDirect Storage, NVIDIA NIXL, and NVIDIA Dynamo—signals that silicon vendors recognize the same shift. Open-sourcing a plugin for the NVIDIA Inference Transfer Library shows WEKA wants widespread adoption, not a walled garden.

OCI’s bare-metal infrastructure with RDMA networking makes it one of the first clouds capable of showcasing the technology without bottlenecks.

This ecosystem convergence—cloud, GPU, and storage—suggests that memory-scaling tech will become a foundational layer of next-gen inference stacks.

Commercial Rollout and What Comes Next

Augmented Memory Grid is now available as a feature for NeuralMesh deployments and listed on the Oracle Cloud Marketplace. Support for additional clouds is coming, though the company hasn’t yet named which.

The implications for AI providers are straightforward:

• Long-context models become affordable to run

• Agentic AI becomes easier to scale and commercialize

• GPU clusters become more efficient

• New monetization models become viable (persistent assistants, multi-user agents, continuous reasoning systems)

WEKA has effectively repositioned memory—from hardware limitation to software-defined superpower.

If compute defined AI’s last decade, memory may define its next one.

Get in touch with our MarTech Experts.

VertexOne, long known for its customer-experience-first approach to utility and energy software, is reorganizing its top bench. The company announced a pair of strategic leadership changes designed to tune up delivery performance and unify the customer journey—a move that reflects how fiercely competitive the utility tech landscape has become.

Energy providers today face more pressure than ever: rising customer expectations, digital modernization mandates, and the operational complexity of distributed energy resources. Vendors in the space aren’t just selling software—they’re selling outcomes. And VertexOne is clearly betting that the right leadership alignment is the lever that drives those outcomes faster.

A New Ops Chief to Tighten the Machine

Keith Ahonen steps into the role of Executive Vice President, Operations, placing him squarely in charge of deployments and delivery across VertexOne’s client portfolio. For utilities, where timelines are tight and integrations are deep, consistency isn’t just nice to have—it's the whole mandate.

Ahonen arrives with 25 years of execution-heavy experience in the energy sector and a recent stint as COO of Accelerated Innovations, which VertexOne acquired in 2024. His task now: streamline internal processes, speed up deployments, and create a delivery organization that scales cleanly as the company grows.

In an industry where system replacements often resemble open-heart surgery for utilities, his focus on reliability and quality isn’t just operational cleanup—it’s a competitive differentiator.

A Chief Client Officer to Own the Entire Journey

While Ahonen sharpens the back end, Tina Santizo takes command of the front. Previously COO, she steps into VertexOne’s newly minted role of Chief Client Officer (CCO). The title signals something clear: VertexOne wants a single leader accountable for the full customer lifecycle, from onboarding to renewals.

It’s a position many tech companies have added in the last few years, especially as cloud vendors compete on lifetime value rather than one-time licensing. For VertexOne, the move formalizes what Santizo has already been known for internally—championing client advocacy and ensuring measurable ROI.

As utilities increasingly evaluate vendors based on delivered value, not just feature checklists, a unified customer-success strategy becomes a powerful retention engine.

Why This Matters for the Utility Tech Market

Across the industry, software vendors are consolidating and optimizing leadership to contend with evolving expectations from utilities. Customers want platforms that adapt quickly, integrate cleanly, and provide clarity on outcomes. VertexOne’s leadership realignment mirrors moves from competitors who are embedding customer success more deeply into product and operations strategy.

This shift also comes at a time when VertexOne is expanding its feature suite, including the recently launched VXconnect—a platform the company has pitched as a “game-changer” for personalized, omnichannel utility customer engagement. Strong operations plus a tightly organized client-experience team could become the backbone that accelerates adoption of such offerings.

The Bigger Theme: Experience is Becoming the Product

Utility software is no longer just about billing engines, outage modules, or portals. Increasingly, CX is the product. Whether a utility chooses Vendor A or Vendor B often comes down to deployment reliability, ongoing guidance, and the confidence that value won’t drop off after go-live.

By elevating ops and client success—two areas where software companies often struggle—VertexOne is signaling that long-term service quality is as central to its strategy as the products themselves.

Looking Ahead

These executive moves won’t instantly transform the company, but they create structural clarity at a time when utilities are demanding more accountability from vendors. With Ahonen refining the delivery engine and Santizo owning the customer journey end-to-end, VertexOne appears to be positioning itself for a market where CX maturity directly influences vendor selection.

The utility tech sector is tightening, expectations are rising, and VertexOne’s reorganization shows it plans to keep pace—not by adding louder marketing claims, but by reinforcing the operational backbone behind them.

Get in touch with our MarTech Experts.

Box and Amazon Web Services have entered a new multi-year strategic collaboration agreement (SCA) aimed at reshaping how enterprises use AI to extract value from content. The deal expands the long-standing partnership between the companies and focuses on developing new Box AI agents powered by AWS infrastructure and foundation models.

The announcement underscores a broader shift: AI agents are moving from prototype to production, and enterprise content is becoming the engine behind them.

Unlocking AI Value from Enterprise Content

According to Box CEO Aaron Levie, the power of AI depends on the context it can access—and that context sits inside documents, contracts, plans, and workflows that drive business operations. Box aims to centralize that intelligence through its Intelligent Content Management (ICM) platform while using AWS as the backbone for scale, security, and compliance.

AWS VP of Agentic AI Swami Sivasubramanian noted that AI agents are redefining how industries operate. The Box collaboration will help organizations securely use their structured and unstructured content as the foundation for agentic workflows.

New AI Agent Integrations with AWS

The partnership introduces a series of deep integrations that expand Box’s AI capabilities and streamline automation across content-heavy workflows. Key additions include:

AI Agents Powered by Amazon Bedrock

New Box AI agents can summarize long documents, generate multi-document FAQs, extract metadata, and trigger automated workflows. Customers can customize these agents using Amazon Bedrock models to fit unique industry or departmental needs.

Multimodal Analysis with Amazon Nova

Using Amazon Nova Multimodal Embeddings, Box AI can analyze text, images, video, and audio together. This unified view improves search accuracy, content intelligence, and automated decision-making across large content repositories.

Amazon Quick Suite Integration

Available today, the new Quick Suite integration lets customers extract insights, generate new files, and act on Box content directly within Quick Suite—boosting productivity for teams handling operational or analytical tasks.

Amazon Q Developer Customization

Developers can use Amazon Q Developer with the Box SDK and self-hosted MCP server to build intelligent apps and automate content workflows faster.

Compatibility with Strands, Kiro, and Bedrock AgentCore

These integrations ensure seamless orchestration between AI agents, connectors, and Box’s ICM platform, enabling secure automation at enterprise scale.

Box to Join AWS Marketplace

A key milestone of the SCA is Box’s upcoming availability in AWS Marketplace in early 2026. This will streamline procurement and deployment for large organizations seeking to centralize spending and accelerate adoption of secure content management and AI-driven workflows.

AWS Marketplace access also strengthens Box’s distribution model, making it easier for regulated industries to buy and deploy Box inside existing cloud environments.

Why This Collaboration Matters

The Box–AWS partnership goes beyond adding AI features. It positions content as a strategic asset and gives enterprises a secure path to deploy agentic workflows without compromising governance.

The combined stack—Box’s ICM platform and AWS’s agentic AI ecosystem—offers scalability, deep compliance, and flexible deployments for industries that handle sensitive data.

Get in touch with our MarTech Experts.

Deepgram has added another milestone to its rapid rise in Voice AI. The company’s enterprise-grade text-to-speech model, Aura-2, has been named a 2025 Customer Experience Innovation Award winner by TMC’s CUSTOMER magazine.

The award highlights companies pushing customer experience forward across every touchpoint—including social channels, automated workflows, and AI-powered agents. And this year, Aura-2 stood out for one reason: it sounds great, but more importantly, it works great.

Built for Work, Not Entertainment

Most TTS models chase entertainment-quality voices. Aura-2 targets the enterprise instead. It is engineered to sound human in the places that matter most—contact centers, regulated workflows, and real-time digital agents.

It provides:

• Domain-specific pronunciation for complex vocabulary
  (drug names, legal terms, identifiers, structured data)

• Sub-200ms TTFB latency, crucial for live voice agents

• Human-like clarity and accuracy

• Pricing that scales for production workloads

The model is powered by Deepgram Enterprise Runtime (DER), which supports deployments across cloud, VPC, and on-prem environments. DER also enables model hot-swapping and real-time optimization, both rare capabilities in the TTS market.

Industry Recognition

TMC CEO Rich Tehrani praised Deepgram for raising the bar on customer experience technology. He highlighted Aura-2 as a model that delivers performance across all customer engagement channels, not just synthetic voice demos.

Deepgram CMO Praveen Rangnath framed Aura-2 as a turning point in enterprise TTS. According to him, the model redefines what production-ready voice AI must deliver—speed, accuracy, consistency, and reliability.

Why Aura-2 Matters

Enterprises are adopting real-time AI agents at unprecedented speed, but most TTS tools still struggle with latency, scaling, and proper pronunciation under load.

Aura-2 directly targets those gaps. Its performance profile makes it suitable for industries where every millisecond and every mispronounced value matters, from customer support to healthcare, fintech, and logistics.

Try Aura-2

Developers can test Aura-2 through a self-serve API, complete with documentation and a real-time playground.

Get in touch with our MarTech Experts.

Marketing measurement has never been easy, but it’s about to get a serious upgrade. Accenture has invested in Alembic, an AI-powered causal intelligence platform built to show which marketing efforts actually generate revenue. The investment comes through Accenture Ventures and includes a strategic partnership designed to push Causal AI deeper into the enterprise stack.

The timing is ideal. According to recent Gartner research, two-thirds of marketing leaders struggle to prove campaign impact. Traditional attribution tools often rely on siloed datasets, lagging models, or incomplete signals. Alembic says it can fix that by grounding measurement in cause-and-effect logic instead of correlation.

A Causal Engine Built for Real-Time Answers

Alembic’s platform ingests data from broadcast channels, social media, site traffic, and direct-to-consumer communications. It then merges those signals with sales data and runs causal analysis to determine what actions drive outcomes. The system assigns an impact score to each channel or marketing event, giving executives a clear view of what moved revenue and why.

The appeal is clear. Marketers want real attribution. Finance teams want accountability. Executives want decisions backed by evidence rather than dashboards that contradict each other.

Accenture CEO Julie Sweet framed the partnership as essential for enterprise transformation. Companies are no longer deploying AI in isolation. They need trusted intelligence at the core of their operations, and Causal AI offers a more reliable foundation than traditional measurement.

Breaking the Limits of Traditional Attribution

Most measurement platforms struggle with data fragmentation. Many cannot handle channels like brand campaigns, event sponsorships, or quick-moving organic social content. Alembic claims its software can analyze those unstructured signals and map the downstream impact even as customer data expands rapidly.

The platform can also model external factors—such as policy changes or unexpected market events—to show how they influence performance. This helps brands adjust spend in real time and stay ahead of shifting conditions.

Alembic CEO Tomás Puig attributes this capability to the company’s NVIDIA SuperPOD compute backbone. The infrastructure gives the platform enough power to run continuous causal calculations and surface insights with minimal delay. “Most companies aren’t short on data,” Puig said. “They’re short on answers.”

A New Measurement Paradigm for the Enterprise

Accenture Song sees the partnership as a turning point for performance measurement. According to Arun Kumar, global customer AI and data lead, Alembic complements methods such as marketing mix modeling but adds the ability to analyze far more variables. Instead of viewing measurement as a post-campaign autopsy, Causal AI turns analytics into a live operational tool.

The partnership also joins a growing ecosystem of AI tools within Accenture Song. Aaru supports strategic planning; Writer enhances content creation; AI Refinery accelerates campaign execution. Alembic slots into the final stage—proving what worked, how it worked, and how to scale it.

Accenture is already piloting Alembic’s technology internally to assess its own marketing initiatives. This early integration signals confidence in the platform and sets the stage for wider client adoption.

Part of a Larger Funding Wave

This investment follows Alembic’s recent Series B round, which was led by Prysm Capital and Accenture. Other participants included Silver Lake Waterman, Liquid 2 Ventures, NextEquity, Friends & Family Capital, and WndrCo. The funding will help Alembic expand its Causal AI engine, enhance its infrastructure footprint, and support a growing roster of enterprise customers.

With demand rising for reliable, real-time attribution, the partnership positions Alembic as a key player in the next phase of AI-driven marketing intelligence. As enterprises look for clarity in a noisy market, Causal AI may prove to be the missing link between massive datasets and actionable decisions.

Get in touch with our MarTech Experts.

Alembic Technologies just pulled off one of the biggest jumps in enterprise AI valuation this year. The Causal AI startup secured $145 million in Series B and growth funding, marking a 15.7x valuation increase since its last round. The investment was led by Prysm Capital and Accenture, signaling strong institutional confidence in a company aiming to redefine how enterprises understand cause and effect.

The move comes at a moment when every B2B vendor claims AI leadership. Yet Alembic’s pitch is sharper: while competitors rely on generic models, it focuses on proprietary data, causal inference, and a compute layer powerful enough to keep up with Fortune 500 demand. That combination has drawn interest from companies such as Delta Air Lines, Mars, NVIDIA, and others seeking clarity in a messy measurement landscape.

A Data Flywheel With Real Competitive Weight

At the core of Alembic’s strategy is a simple idea with major implications: better insights create better decisions, which then generate better data. This closed loop forms what CEO Tomás Puig calls a compounding data flywheel, and it’s difficult for rivals to replicate.

Many enterprises are drowning in dashboards and correlation-based reporting. However, Alembic claims its Causal AI identifies what actually drives outcomes. That advantage positions the company as a key intelligence layer rather than another analytics feature.

Accenture’s involvement reinforces that view. The consulting firm will integrate Alembic’s Causal engine into its enterprise transformation work, giving clients tools that move beyond pattern recognition toward evidence-based decisions. As Accenture CEO Julie Sweet put it, large companies need “verifiable, cause-and-effect insights” to act quickly and safely.

Big Customers, Bigger Impact

Enterprise customers are already using the platform to quantify decisions that were previously guesswork. Delta Air Lines linked Olympic sponsorship activations to ticket sales within days. Mars measured the exact dollar impact of viral brand moments. North Sails optimized digital spend for record returns. One Fortune 500 company grew its sales pipeline by 37% with precise attribution.

These examples illustrate why Alembic is drawing attention. Most organizations are not short on data; they are short on certainty. Causal AI promises to close that gap.

A Supercomputer Strategy Built for the Long Game

Alembic is pairing its funding with serious infrastructure investment. The company will deploy a new NVIDIA NVL72 superPOD cluster at Equinix’s SV11 data center, running NVIDIA AI Enterprise across its stack. This system is engineered for spiking neural networks, high-speed graph processing, and continuous-learning workloads.

This will be the company’s second dedicated supercomputing cluster, creating bi-coastal redundancy and ensuring the compute headroom needed for real-time causal analysis. Alembic positions this as a strategic moat. Instead of relying on shared cloud resources, it will operate a private AI fleet optimized for its unique workloads.

The deeper tie to NVIDIA also reflects a broader trend: enterprise AI platforms increasingly require custom compute infrastructure to maintain speed, security, and differentiation.

Defining the Next Layer of Enterprise Intelligence

Investors see Alembic as more than a vertical analytics solution. Many describe it as an emerging foundational model for enterprise decision-making. Instead of generating text or images, it generates causal truth, and brands are paying attention.

Prysm Capital’s team sees the company as a “mission-critical intelligence layer,” while WndrCo partners highlight the platform’s ability to deliver what marketers have sought for decades: clear, quantifiable attribution that informs where every dollar should go.

The company is also building a reputation for marrying deep research with commercial relevance. Its approach combines spiking neural networks, advanced graph modeling, real-time simulation, and high-performance compute into a system that updates continuously as new data enters the ecosystem.

The Market Impact

With this round, Alembic is positioned to influence several fast-shifting categories:

• Marketing measurement, where correlation-based models are losing credibility

• AI-driven budgeting, as brands face pressure for provable ROI

• Enterprise intelligence platforms, which increasingly compete on proprietary data

• Custom AI infrastructure, especially among companies needing guaranteed compute

Causal AI sits at the intersection of all four. That gives Alembic a strategic lane with few direct rivals and strong tailwinds as enterprises rebuild their data strategies around reliability rather than volume.

The Bottom Line

Alembic now has the capital, compute, and customer base to accelerate its push into enterprise AI. The company’s focus on Causal intelligence sets it apart in a market dominated by generative hype. With a new superPOD, deeper ties to Accenture, and accelerating Fortune 500 adoption, it is shaping a category that could define the next era of decision intelligence.

If Alembic delivers on its promise, the industry may soon shift from asking what happened to understanding why—and acting with far more confidence.

Get in touch with our MarTech Experts.

ON24 just added another weapon to its engagement stack. The company secured U.S. Patent No. 12,445,698 B2 for its AI engine that finds and extracts “Key Moments” from long-form video content. The system works across webinars, virtual events, digital conferences, and recorded demos.

The feature solves a long-standing marketing headache. Teams often run events that generate hours of footage but struggle to reuse that content. ON24’s patented technology tackles that gap by automating the discovery of highlight moments. The platform then packages them into short clips that can slip into nurture streams, partner enablement programs, landing pages, and social feeds.

Turning Engagement Behavior Into Short-Form Assets

The AI engine does more than trim videos. It examines engagement signals from the event itself. That includes viewing behavior, interactions, and content consumption patterns. The model then ranks moments that matter and extracts them into short, high-value videos.

This helps B2B marketers extend the shelf life of every event. Instead of a single live session, teams can quickly launch follow-ups, repurposed assets, and targeted campaigns—without manual editing or guesswork.

According to ON24 CEO Sharat Sharan, the core of the company’s strategy is AI-driven engagement. He says the new patent strengthens ON24’s position as an AI-enabled platform that understands audience behavior in real time. It also helps the company scale personalized content across channels.

Built by Engineers and Data Scientists for Demand Teams

The “Key Moments” engine is the product of ON24 developers and machine learning experts. The goal was simple: pull out the most engaging sections from any event. That includes product walkthroughs, customer stories, expert analysis, or Q&A sessions.

ON24 says the tech benefits a wide range of teams—marketing, customer success, sales, partner management, and training groups. Each can reuse “Key Moments” to reinforce messaging or support education workflows.

CTO Jayesh Sahasi says the patent advances ON24’s vision of connected, AI-led engagement. The system ties together behavioral data, automation, and content analysis. It lets teams transform a single event into continuous, data-backed campaigns.

Why This Matters for the B2B Engagement Market

Short-form content is the new currency in B2B marketing. Marketers want rapid content delivery, personalized outreach, and scalable video workflows. ON24’s patent supports that shift by automating one of the slowest steps in the process—manual video review.

Competitors offer clip-generation tools, but ON24’s differentiator is its use of engagement data tied to event behavior. It’s not guessing what content works; it’s identifying what audiences respond to.

As AI becomes central to B2B engagement platforms, this patent strengthens ON24’s position in a crowded market. It also signals the company’s larger push toward AI-powered orchestration across the full customer journey.

Get in touch with our MarTech Experts.

AI has already transformed market research, but according to the new 2026 Market Research Trends report from Qualtrics, the divide between teams using basic AI and those embracing purpose-built capabilities is widening rapidly. The stakes aren’t small: research groups relying only on generic tools are four times more likely to lose influence inside their organizations.

Meanwhile, 72% of teams using synthetic responses, agentic AI, and AI-native workflows say their organizations rely on research far more than they did last year—momentum that’s translating directly into higher budgets. Traditional teams, however, are almost twice as likely to face stagnant or shrinking demand for their work.

“In today’s fast-moving economies, rapid access to consumer insights is a huge advantage,” said Ali Henriques, Executive Director of Edge at Qualtrics. “The teams embracing AI are reimagining what research looks like, asking bigger questions, and moving earlier in the innovation cycle.”

Purpose-Built AI Becomes the New Standard

AI adoption has crossed a maturity threshold. More than half of researchers now use AI regularly, and nearly nine in ten have experimented with it, but the report shows a clear shift away from generic chatbots toward AI embedded directly in research platforms. Purpose-built capabilities grew from 62% to 66% adoption, while usage of general-purpose tools dropped by nearly ten points.

The teams gaining the most traction are those leaning into specialist functions. Conversational analytics and visual content analysis—both at 49% adoption—give researchers deeper qualitative insight at a fraction of the time. What once took weeks can now be processed in hours.

Synthetic data is driving an even more dramatic evolution. Researchers using synthetic datasets are:

• 11% more likely to engage in early-stage innovation

• 7% more likely to run go-to-market studies

• 5% more likely to perform final product testing

Among those who’ve adopted it, 45% now consider synthetic data their most reliable source, surpassing traditional online panels—a remarkable shift for an industry built on human surveys.

Brands like Gabb are already using Qualtrics’ purpose-built synthetic model to reduce fielding costs, accelerate discovery, and test messaging against emerging trends. As Research Director Garred Sheppard described it, “Synthetic data became our cultural radar—cutting timelines from a week to hours while letting us validate high-stakes decisions with human panels.”

AI Agents Push Research Toward End-to-End Automation

Another major shift is the rise of agentic AI. While only 15% of researchers use AI agents today, nearly 80% expect that these tools will handle more than half of research projects end-to-end within the next three years.

Efficiency gains are already visible. Among teams using agentic AI, 84% report significantly higher efficiency, compared with 68% of those who haven’t tried it.

Henriques said the biggest unlock isn’t workload reduction—it’s democratization. Product teams can test ideas without submitting requests. Marketing can evaluate sentiment without waiting on insights teams. Executives can explore new markets directly. “The barrier to insights is no longer specialist knowledge,” she said. “It’s simply asking the right question.”

Leadership vs. Frontline: A Growing Execution Gap

Despite major investments in AI, many organizations aren’t seeing the full return. The report highlights a sharp misalignment between research leaders and individual contributors:

• 39% of leaders say AI has revolutionized their processes vs. 19% of frontline researchers

• Only 5% of leaders fear layoffs due to AI vs. 15% of individual contributors

• 68% of leaders consider themselves synthetic data experts vs. 41% of contributors

• 79% of leaders trust synthetic data quality vs. 61% of contributors

This mismatch results in underused tools, wasted budget, and slower execution—while competitors with tighter alignment surge ahead.

“When frontline teams don’t buy in, expensive AI tools go unused,” Henriques warned. “Organizations need shared definitions of success, real hands-on training, and clarity across levels about the practical applications of new AI capabilities.”

A Global Snapshot of a Rapidly Changing Industry

The findings come from a global Qualtrics study conducted in Q3 2025 with more than 3,000 research professionals across 14 countries. The data reveals a sector in transition—from manual workflows and traditional surveys to hybrid human-synthetic models, autonomous research agents, and a new definition of what strategic research teams look like.

The message from Qualtrics is clear: the teams that invest in purpose-built AI now will set the pace for the next decade of research. The ones that don’t risk being left behind.

Get in touch with our MarTech Experts.