Connect with us

Technology

Dyna Robotics unveils DYNA-2 World-Action Model, demonstrating first true scaling law in robotics powered entirely by human data

Published

on

DYNA-2 is the first robot foundation model pre-trained on over 1 millions hours of egocentric human video – roughly 170 years of continuous waking experience – and establishes a new scaling axis for physical AI by training on pure video instead of scarce action data.

Groundbreaking cross-embodiment transfer across stationary robot arms, humanoid prototypes, and dexterous robotic hands eliminates training bottlenecks and paves a clear path toward 10 million hours of training data – enabling robots to master new physical tasks with just hours of local fine-tuning and delivering zero-shot production-level performance.

REDWOOD CITY, Calif., Aug. 10, 2026 /PRNewswire/ — Dyna Robotics today announced DYNA-2, its flagship World-Action Model (WAM) and the first robot foundation model trained on over 1 million hours of data – equivalent to 170 years of continuous waking experience.

Powered entirely by human egocentric video, DYNA-2 is based on foundational research that demonstrates the first human-to-robot scaling law for robotics, proving that robot performance improves smoothly and predictably with every added hour of human data, without hitting plateaus. This is a paradigm shift for robotic foundation models; by training on video data alone rather than scarce action data or manual teleoperation, Dyna Robotics has discovered a new scaling axis for physical AI, showing that robotics can bypass its greatest bottleneck – scarce, hand-collected teleoperation data – by using video datasets to predictably scale capabilities.

Dyna Robotics’ robots, powered by its previous DYNA-1 model, are already deployed in production in high-demand environments like hotels, restaurants, and laundromats. DYNA-2 is a major step forward in making production-grade, autonomous robots commercially-viable. In high-precision manufacturing tasks, DYNA-2 increased task success rates from 20% to 80–90% purely through pre-training scale, without requiring any changes to post-training data. DYNA-2 enables zero-shot performance at production-level speeds and quality across new deployment sites.

Action data has long choked generalist robotics; collecting physical teleoperation data manually is expensive, slow, and unscalable. DYNA-2 bypasses this bottleneck by establishing a cross-embodiment transfer scaling law. Pre-trained on 1 million hours of human video using a dual next-frame and next-action world-modeling architecture, DYNA-2 enables physical intuition and spatial reasoning gained from human motion to transfer directly to robot hardware, even without seeing a single frame of robot data in pre-training. This dramatically lowers the barrier to training robots on new tasks.

“For years, generalist robotics has been choked by a data bottleneck: collecting physical teleoperation data manually simply cannot scale to general intelligence,” said Dyna Robotics co-founder Jason Ma. “Action data is scarce, but video is everywhere, and with DYNA-2, we showed that physical intuition doesn’t require millions of hours of training on a robot arm – it can be learned directly from human video. By building a World-Action Model that imagines how the physical world moves before taking action, we give robots spatial reasoning and contact physics that traditional vision-language models simply lack and show that physical AI can scale smoothly and predictably, unlocking commercial-grade automation far faster than previously thought possible.”

Production-grade performance in hours

DYNA-2 can be quickly adapted to perform tasks on many different robot platforms with just a few hours of local fine-tuning:

Aggregated over 15 benchmark tasks, DYNA-2 policies pre-trained with more human data consistently performed betterIn one case, with just 13 minutes of data, DYNA-2 could command a pair of five-fingered robot hands to twist open a bottle cap

Significant performance increases versus Vision-Language-Action models

DYNA-2 shifts the foundation model architecture from Vision-Language-Action (VLA) adaptations to a native WAM built on video generation. In head-to-head physical evaluations against DYNA-1, Dyna Robotics’ VLA model, under matched training steps and datasets, DYNA-2 demonstrated decisive control and task resilience advantages:

DYNA-2 successfully completed tasks 1.55 times more than DYNA-1 in real-world evaluations. At a customer deployment, DYNA-2 achieved 87% pass rate while DYNA-1 achieved only 46% pass rate.DYNA-2 is able to recover from physical disturbances without human intervention when completing tasks requiring dexterous manipulation like chopping food and clearing workspaces, compared to the VLA baseline which failed in a single loop and required manual intervention to recover.DYNA-2’s video co-training algorithm improved scores by 133% on instruction-following tasks where robots must perform distinct motions based on user-provided commands

About Dyna Robotics
Dyna Robotics makes general-purpose robots powered by a proprietary embodied AI foundation model that generalizes and self-improves across varied environments with commercial-grade performance. Dyna’s robots have been deployed at customers across multiple industries. Dyna Robotics was founded by repeat founders Lindon Gao and York Yang, who sold Caper AI for $350 million, and former DeepMind research scientist Jason Ma. The company is backed by top investors, including CRV and First Round. Learn more at dyna.co. 

View original content to download multimedia:https://www.prnewswire.com/news-releases/dyna-robotics-unveils-dyna-2-world-action-model-demonstrating-first-true-scaling-law-in-robotics-powered-entirely-by-human-data-302847114.html

SOURCE DYNA Robotics

Continue Reading

Technology

Allooloo Tops 100M Capital Markets AI Agent Traffic a Month

Published

on

By

Agentic datasets on every listed company in 34 countries: sourced, dated and signed.

VANCOUVER, BC and LONDON, Nov. 20, 2026 /PRNewswire/ — Allooloo Technologies Corp. (Allooloo), maker of agentic datasets for capital markets artificial intelligence (AI) agents, today reported 100M AI agent traffic a month. That is 39 calls a second, 3.4 million a day, across getissuer.ai and Issuerwire.ai, running on 24 Microsoft Azure hyperscale locations globally with always-on Amazon Web Services (AWS) redundancy.

The Data Readiness Gap

Investment banks are not short of AI models. They are short of data an AI agent can use. In a 2026 Teradata study, 77% of technology leaders said 20% or less of their enterprise data is ready for AI agents. Banks are closing that gap on the data they hold, with one governed way in and the source, the date and the owner on every record.

The data a bank does not hold is the other half. An AI agent working on a security needs the issuer, its disclosures and whether it may be ordered, in every market the bank trades. Allooloo delivers that half agent-ready: 41.9K listed companies in 34 countries, a source and a date on every field, and a signed receipt for every read, over Model Context Protocol (MCP).

Two Products

Know Your Product (KYP) is the duty to know a security before it is recommended or sold. Allooloo KYP, at getissuer.ai, answers it for the AI agent: who the issuer is, whether it may be ordered, and who it is for. It covers 41.9K listed companies on 79 exchanges in 34 countries.

Issuerwire.ai is the dated disclosure wire: 3.1 million dated disclosures across 34 countries, each with its source.

Who it matters to:

Investment bank or dealer: its AI agents can show where each fact came from.The issuer: every AI agent reads the same answer about it.Compliance: a signed receipt for every cleared read.

A Signed Receipt for Every Read

Every cleared read on Allooloo leaves a signed receipt, the Record of Cleared Orders (RCO): what was read, when, and under which market’s rules.

Each Market in Its Own Country

Each market is kept in its own country or region, across 24 Microsoft Azure regions, on one MCP server per market. A second copy of every market is made each night on AWS and locked against deletion for 365 days. Allooloo is a Microsoft AI Cloud Partner.

Around the Clock

The products are made around the clock by Allooloo’s own team and AI agents, on the Allooloo Algo. AI agents call them around the clock and pay per call, with no seat and no licence.

“Today we are announcing a milestone: 100M AI agent traffic a month,” said Matthew Keddy, Chief Executive Officer of Allooloo Technologies Corp. “These are the AI agents of capital markets participants, and they want crisp data, five years deep, on 41.9K listed companies across 79 exchanges: sourced, dated and signed. Gain the edge with us.”

In Session and Enterprise Grade

Allooloo is sold two ways. In session, an AI agent calls Allooloo mid-task and pays as it goes (PayGo) on x402, the open payment standard for AI agents, with no seat and no licence. At enterprise grade, a firm licenses the datasets to design, build and run its own AI agents.

About Allooloo Technologies Corp.

Allooloo Technologies Corp. makes agentic capital markets data products for the AI agents of banks, dealers and issuers. Its products, getissuer.ai and Issuerwire.ai, cover 41.9K listed companies on 79 exchanges in 34 countries, kept across 24 Microsoft Azure regions with on-the-fly redundancy on Amazon Web Services (AWS). The result is always-on agentic datasets: in session for AI lab users, and enterprise grade for customers who design, build and run their own AI agents. Allooloo is headquartered in Vancouver, Canada. Please visit: allooloo.ai 

View original content:https://www.prnewswire.com/news-releases/allooloo-tops-100m-capital-markets-ai-agent-traffic-a-month-302947422.html

SOURCE Allooloo Technologies Corp.

Continue Reading

Technology

Allooloo Tops 100M Capital Markets AI Agent Traffic a Month

Published

on

By

Agentic datasets on every listed company in 34 countries: sourced, dated and signed.

VANCOUVER, BC and LONDON, Nov. 21, 2026 /PRNewswire/ — Allooloo Technologies Corp. (Allooloo), maker of agentic datasets for capital markets artificial intelligence (AI) agents, today reported 100M AI agent traffic a month. That is 39 calls a second, 3.4 million a day, across getissuer.ai and Issuerwire.ai, running on 24 Microsoft Azure hyperscale locations globally with always-on Amazon Web Services (AWS) redundancy.

The Data Readiness Gap

Investment banks are not short of AI models. They are short of data an AI agent can use. In a 2026 Teradata study, 77% of technology leaders said 20% or less of their enterprise data is ready for AI agents. Banks are closing that gap on the data they hold, with one governed way in and the source, the date and the owner on every record.

The data a bank does not hold is the other half. An AI agent working on a security needs the issuer, its disclosures and whether it may be ordered, in every market the bank trades. Allooloo delivers that half agent-ready: 41.9K listed companies in 34 countries, a source and a date on every field, and a signed receipt for every read, over Model Context Protocol (MCP).

Two Products

Know Your Product (KYP) is the duty to know a security before it is recommended or sold. Allooloo KYP, at getissuer.ai, answers it for the AI agent: who the issuer is, whether it may be ordered, and who it is for. It covers 41.9K listed companies on 79 exchanges in 34 countries.

Issuerwire.ai is the dated disclosure wire: 3.1 million dated disclosures across 34 countries, each with its source.

Who it matters to:

Investment bank or dealer: its AI agents can show where each fact came from.The issuer: every AI agent reads the same answer about it.Compliance: a signed receipt for every cleared read.

A Signed Receipt for Every Read

Every cleared read on Allooloo leaves a signed receipt, the Record of Cleared Orders (RCO): what was read, when, and under which market’s rules.

Each Market in Its Own Country

Each market is kept in its own country or region, across 24 Microsoft Azure regions, on one MCP server per market. A second copy of every market is made each night on AWS and locked against deletion for 365 days. Allooloo is a Microsoft AI Cloud Partner.

Around the Clock

The products are made around the clock by Allooloo’s own team and AI agents, on the Allooloo Algo. AI agents call them around the clock and pay per call, with no seat and no licence.

“Today we are announcing a milestone: 100M AI agent traffic a month,” said Matthew Keddy, Chief Executive Officer of Allooloo Technologies Corp. “These are the AI agents of capital markets participants, and they want crisp data, five years deep, on 41.9K listed companies across 79 exchanges: sourced, dated and signed. Gain the edge with us.”

In Session and Enterprise Grade

Allooloo is sold two ways. In session, an AI agent calls Allooloo mid-task and pays as it goes (PayGo) on x402, the open payment standard for AI agents, with no seat and no licence. At enterprise grade, a firm licenses the datasets to design, build and run its own AI agents.

About Allooloo Technologies Corp.

Allooloo Technologies Corp. makes agentic capital markets data products for the AI agents of banks, dealers and issuers. Its products, getissuer.ai and Issuerwire.ai, cover 41.9K listed companies on 79 exchanges in 34 countries, kept across 24 Microsoft Azure regions with on-the-fly redundancy on Amazon Web Services (AWS). The result is always-on agentic datasets: in session for AI lab users, and enterprise grade for customers who design, build and run their own AI agents. Allooloo is headquartered in Vancouver, Canada. Please visit: allooloo.ai 

Contact: Matthew Keddy, CEO. mk@allooloo.ai +1 647 237 7744

View original content:https://www.prnewswire.com/apac/news-releases/allooloo-tops-100m-capital-markets-ai-agent-traffic-a-month-302947427.html

SOURCE Allooloo Technologies Corp.

Continue Reading

Technology

Ziehm-Orthoscan Announces Executive Leadership Transition Following 18 Years of Significant Growth

Published

on

By

Ziehm-Orthoscan Announces CEO Transition and Executive Leadership Appointments

SCOTTSDALE, Ariz., Nov. 20, 2026 /PRNewswire-PRWeb/ — Ziehm-Orthoscan Inc., the leader in mobile C-arm imaging solutions, today announced an executive leadership transition as President and Chief Executive Officer Nelson Mendes prepares to step down on December 31, 2026, following 18 years of distinguished leadership.

Effective January 1, 2027, Travis Prikryl will assume the role of Chief Executive Officer, and Hogie Saunders will be appointed Chief Operating Officer. Both executives will also serve as Managing Directors of Ziehm-Orthoscan Inc.

Under Mendes’ leadership, Ziehm-Orthoscan experienced extraordinary growth and transformation, evolving from a small U.S. operation into a major contributor to the company’s global business. Since 2009, U.S. operations have grown from approximately six employees to nearly 200, reflecting Ziehm-Orthoscan’s significant expansion and continued investment. This growth includes the establishment of a new headquarters and production facility in Scottsdale, Arizona, which serves as the global center for the development and manufacturing of all Orthoscan mini C-arm systems.

A significant milestone in this transformation was the acquisition and integration of Orthoscan in 2015, bringing together Ziehm’s full-size mobile C-arm technology and Orthoscan’s mini C-arm expertise. This strategic expansion established a comprehensive mobile imaging portfolio, strengthened the company’s competitive position, and accelerated its growth throughout the U.S. healthcare market.

Throughout his tenure, Mendes championed innovation, operational excellence, and customer relationships, helping establish Ziehm-Orthoscan as a trusted imaging partner to healthcare institutions nationwide.

To ensure a seamless transition, Mendes will remain with the company as a full-time strategic advisor through March 31, 2027, followed by a part-time consulting role supporting the organization’s continued development.

The appointments of Travis Prikryl and Hogie Saunders represent the next chapter in Ziehm-Orthoscan’s long-term growth strategy. Together with the company’s experienced management team, they will focus on advancing imaging innovation, strengthening customer partnerships, expanding market opportunities, and building on the strong foundation established over the past 18 years.

Ziehm-Orthoscan extends its sincere appreciation to Nelson Mendes for his exceptional leadership and contributions while expressing full confidence in the incoming executive team to guide the organization into its next phase of growth and success.

About Ziehm-Orthoscan

Ziehm-Orthoscan is the leader of innovative mobile imaging solutions, combining Ziehm’s expertise in full-size mobile C-arms with Orthoscan’s specialized mini C-arm technology. The company is committed to advancing medical imaging through innovation, exceptional service, and solutions designed to meet the evolving needs of healthcare professionals.

Media Contact

Rich Shaul, Vice President Marketing, Ziehm-Orthoscan, Inc., 1 4805038010, marketing@ziehm-orthoscan.com, https://ziehm-orthoscan.com/

View original content to download multimedia:https://www.prweb.com/releases/ziehm-orthoscan-announces-executive-leadership-transition-following-18-years-of-significant-growth-302946604.html

SOURCE Ziehm-Orthoscan, Inc.

Continue Reading

Trending