Snowflake Unveils AI Data Cloud Innovations at Summit 2025
Snowflake (NYSE: SNOW), the AI Data Cloud company, announced major product innovations at its annual user conference, Snowflake Summit 2025. These updates are set to transform how enterprises manage, analyze, and activate data in the AI era.
The company revealed enhancements across data engineering, compute performance, analytics, and agentic AI. These innovations aim to eliminate data silos and connect enterprise data to business action. At the same time, they maintain control, simplicity, and governance.
James Petter, Vice President, Snowflake EMEA, stated that the new updates redefine what organizations can expect from a modern data platform. He emphasized that the company's goal is to make AI and machine learning workflows more accessible, trusted, and efficient.
Snowflake introduced Snowflake Openflow, a multi-modal data ingestion service now generally available on AWS. It enables users to connect to nearly any data source and drive value from any architecture. Openflow removes data fragmentation by unifying formats and systems.
The service is powered by Apache NiFi™, automating data flow between systems. It allows data engineers to create custom connectors in minutes, running them on Snowflake's managed platform. It supports a wide range of sources like Box, Google Ads, Proofpoint, ServiceNow, Workday, and Zendesk, among others.
Key capabilities include: Hundreds of ready-to-use connectors
Integration with cloud object stores and messaging platforms
Snowflake also revealed new compute innovations. These include Standard Warehouse Generation 2 (Gen2), now generally available, which offers 2.1x faster performance. Another addition, Adaptive Compute, is now in private preview. This feature automatically sizes and shares resources for better performance and lower costs.
The company reported the upcoming release of Snowflake Intelligence and Cortex Agents, both in public preview soon. These tools enable users to ask natural language questions and get insights from structured and unstructured data. Powered by models from Anthropic and OpenAI, they run securely within Snowflake.
Another announcement was the Data Science Agent, now in private preview. It helps data scientists by automating tasks like data preparation, feature engineering, and model training using Anthropic's Claude.
According to Snowflake, more than 5,200 customers, including BlackRock, Luminate, and Penske Logistics, are already using Cortex AI to transform their operations.
The company also introduced SnowConvert AI and Cortex AISQL. These tools support fast and cost-effective migration from legacy systems and enable generative AI-powered SQL analytics. Both are designed for high performance and efficiency.
Additionally, Snowflake revealed updates to its Marketplace. New agentic products like Cortex Knowledge Extensions will soon be available. These allow enterprises to enrich AI agents with third-party data while ensuring data protection and attribution. Users can access content from The Associated Press and other providers.
Through these developments, Snowflake aims to empower global organizations to modernize their data strategies with enterprise-ready AI.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
TECHx
3 hours ago
- TECHx
AI Servers Drive META Region's 2025 Tech Boom
Home » Tech Market Insights » AI Servers Drive META Region's 2025 Tech Boom The Middle East, Turkey, and Africa (META) region is witnessing rapid growth in AI infrastructure, with AI servers leading revenue gains in early 2025. According to CONTEXT's SalesWatch Distribution report, AI-capable servers are now the main driver of tech-related revenue across the region. This expansion highlights a growing focus on AI data center investments, though progress remains uneven across different countries and infrastructure categories. The Ascendancy of AI-Capable Servers AI servers accounted for 42% of total server revenues in Q2 2025. This is a dramatic jump from 21% in Q4 2024 and 18% in Q1 2025. The upward trend reflects rising demand for AI-optimized server platforms, with April and May 2025 investments surpassing $50 million. Combined server revenues in those two months edged past those of Q2 2024. From January to May 2025, overall server revenues in META jumped 66%, with May showing a 106% year-on-year increase. The rise is largely fueled by AI infrastructure projects, especially in the Middle East. Divergent Trends in Enterprise Networking and Storage However, this momentum does not extend equally to all tech infrastructure areas. Enterprise networking and storage showed weaker performance. The enterprise networking category, which includes business switches, routers, wireless devices, and software, saw a 21% revenue drop from January to May 2025 versus the same period last year. Network switching took the hardest hit, with a 25% decline in switch sales and expansion modules. Wireless access points performed slightly better but still ended May with a 25% drop compared to May 2024. This decline aligns with past volatility in networking equipment sales, affected by supply chain issues, price hikes, and changes in office space needs due to hybrid work. However, future gains are expected with Wi-Fi 7 upgrades and AI-led data center growth. The enterprise storage segment, which covers storage area networks, HCI systems, and related software, had a flat start to 2025. There was zero growth compared to 2024 in the first five months. In May alone, revenues dipped 28% year-on-year. The hyper-converged infrastructure (HCI) sub-category was the main drag, falling 22%, while traditional arrays held steady due to a strong flash storage segment. Still, demand is likely to rise, as AI workloads require higher-capacity storage and faster bandwidth solutions. Geographic Variations in AI Investment The United Arab Emirates (UAE) leads AI investment in the META region. The country hosts major projects like Stargate UAE, a 1-gigawatt AI cluster in Abu Dhabi. It targets AI applications across government, energy, healthcare, education, and transport. The UAE was also the first to give its entire population access to ChatGPT, showcasing its national commitment to AI. Meanwhile, Saudi Arabia is seeing gains too. Its AI server investments drove a 28% rise in server revenue in 2025 over 2024. Turkey is also ramping up, with growing UAE-Turkey collaborations, such as the upcoming 100-megawatt AI-capable data center in Ankara by UAE-based Khazna. Uneven Growth Across Countries and Segments In enterprise networking, South Africa and the UAE saw smaller revenue drops of just 5% compared to 2024. But Turkey and Saudi Arabia faced sharper declines of about 30% year-on-year. This shows a divided growth pattern across the region's infrastructure sectors. AI Servers Are Reshaping META's Tech Future So far, 2025 has marked a strong start for infrastructure spending in the META region, especially compared to the past two years. AI-driven server sales are the biggest success story, indicating a major tech shift. However, the benefits are not evenly spread. As AI servers gain ground, future growth will depend on how well countries and companies adapt their investments across storage, networking, and data center infrastructure.
Web Release
3 hours ago
- Web Release
Lilly's once-weekly insulin efsitora alfa demonstrated A1C reduction and a safety profile consistent with daily insulin in multiple Phase 3 trials
Eli Lilly and Company (NYSE: LLY) today announced detailed results from QWINT-1, QWINT-3, and QWINT-4 Phase 3 clinical trials evaluating the safety and efficacy of investigational once-weekly insulin efsitora alfa (efsitora) in adults with type 2 diabetes who used insulin for the first time, previously used daily basal insulin, and previously used daily basal insulin and mealtime insulin, respectively. In each trial, once-weekly efsitora met the primary endpoint of non-inferior A1C reduction compared to daily basal insulin. The complete results from these studies were presented at the American Diabetes Association (ADA) 85th Scientific Sessions. In QWINT-1, efsitora reduced A1C by 1.31% compared to 1.27% for insulin glargine at week 52 for the efficacy estimand.1,2 In the trial, efsitora was titrated to four fixed doses at four-week intervals, as needed for blood glucose control.3 In QWINT-3, efsitora reduced A1C by 0.86% compared to 0.75% for insulin degludec at week 26 for the efficacy estimand.4 In QWINT-4, efsitora reduced A1C by 1.07% compared to 1.07% for insulin glargine at week 26 for the efficacy estimand.5 In these two trials, efsitora was administered using traditional insulin dosing with adjustments based on each patient's glucose level. 'The novel fixed-dose regimen used in QWINT-1 for once-weekly efsitora, which consisted of only four single-dose titration options, has the potential to facilitate and simplify insulin therapy, reducing the hesitation often associated with starting insulin to treat type 2 diabetes,' said Dr. Julio Rosenstock, senior scientific advisor for Velocity Clinical Research at Medical City Dallas, clinical professor of medicine, University of Texas Southwestern Medical Center, and lead trial investigator for QWINT-1. 'A simpler, once-weekly regimen with efsitora may help people with type 2 diabetes initiate and manage insulin therapy with the goal of improving blood sugar levels. Across all QWINT trials, the results showed that once-weekly efsitora controlled glucose as effectively as the most popular once-daily basal insulins.' QWINT-1 Primary Endpoint QWINT-3 Primary and Key Secondary Endpoints QWINT-4 Primary and Key Secondary Endpoints Efficacy Estimand Treatment-Regimen Estimand6 Primary Endpoint – A1C Reduction (Resulting A1C) at Week 52 Efsitora -1.31% (6.92%) -1.19% (7.05%) Glargine -1.27% (6.96%) -1.16% (7.08%) Efficacy Estimand Treatment-Regimen Estimand Primary Endpoint – A1C Reduction (Resulting A1C) at Week 26 Efsitora -0.86% (6.93%) -0.81% (6.99%) Degludec -0.75% (7.03%) -0.72% (7.08%) Key Secondary Endpoint – Rates of Clinically Significant or Severe Nocturnal Hypoglycemic Events Per Patient-Year of Exposure up to Week 787,8 Efsitora 0.11 Degludec 0.10 Key Secondary Endpoint – Percent Time in Range (70-180 mg/dL) During the 4 Weeks Prior to Week 26 Efsitora 62.8% 61.4% Degludec 61.3% 61.0% Efficacy Estimand Treatment-Regimen Estimand Primary Endpoint – A1C Reduction (Resulting A1C) at Week 26 Efsitora -1.07% (7.12%) -1.01% (7.17%) Glargine -1.07% (7.11%) -1.00% (7.18%) Key Secondary Endpoint – Participants Achieving A1C <7% at Week 26 Without Nocturnal Hypoglycemia Efsitora 39.5% 38.6% Glargine 36.6% 35.9% Key Secondary Endpoint – Rates of Clinically Significant or Severe Nocturnal Hypoglycemic Events Per Patient-Year of Exposure up to Week 26 Efsitora 0.67 Glargine 1.00 'Building on Lilly's legacy of innovation in insulin therapy, once-weekly efsitora may offer a significant advancement for people with type 2 diabetes who need insulin by eliminating over 300 injections per year,' said Jeff Emmick, MD, Ph.D., senior vice president of product development at Lilly. 'These results reinforce the potential for once-weekly efsitora to help reduce the overall burden of insulin therapy through a simplified treatment approach. We look forward to working with regulatory agencies to bring this innovation to patients around the world.' Across the three trials, efsitora demonstrated an overall safety profile similar to two of the most commonly used daily basal insulin therapies for the treatment of type 2 diabetes. In QWINT-1, efsitora resulted in approximately 40% fewer hypoglycemic events compared to insulin glargine, with estimated combined rates of severe or clinically significant hypoglycemic events per patient-year of exposure of 0.50 with efsitora vs. 0.88 with insulin glargine at 52 weeks. In QWINT-3, these rates were 0.84 with efsitora vs. 0.74 with insulin degludec at 78 weeks. In QWINT-4, rates of estimated combined rates of severe or clinically significant hypoglycemic events per patient-year of exposure were 6.6 with efsitora vs. 5.9 with insulin glargine at 26 weeks. Lilly plans to submit efsitora for the treatment of adults with type 2 diabetes to global regulatory agencies by the end of this year. About the QWINT clinical trial program The QWINT Phase 3 global clinical development program for insulin efsitora alfa (efsitora) in diabetes began in 2022 and has enrolled more than 3,000 people living with type 2 diabetes across four global registration studies. QWINT-1 (NCT05662332) was a parallel-design, open-label, treat-to-target, randomized controlled clinical trial comparing the efficacy and safety of efsitora as a once-weekly basal insulin using a fixed dose escalation to daily insulin glargine for 52 weeks in insulin-naïve adults with type 2 diabetes. The trial randomized 795 participants across the U.S., Argentina and Mexico to receive efsitora once weekly or insulin glargine once daily, administered subcutaneously. Participants treated with efsitora received a starting dose of 100 units of insulin, followed by escalation to fixed dosages of 150 units, 250 units and 400 units every 4 weeks, as needed, until achieving a target fasting blood glucose of 80-130 mg/dL. Participants with fasting blood glucose greater than 130 mg/dL on or after 16 weeks were transferred to flexible dosing. The primary objective of the trial was to demonstrate non-inferiority in reducing A1C at week 52 with efsitora compared to daily use of insulin glargine. QWINT-3 (NCT05275400) was a multicenter, randomized, parallel-design, open-label trial comparing the efficacy and safety of efsitora as a once-weekly basal insulin to insulin degludec for 78 weeks after a three-week lead-in followed by a five-week safety follow up period, in adults with type 2 diabetes who are currently treated with basal insulin. The trial randomized 986 participants across the U.S., Argentina, Hungary, Japan, Korea, Poland, Puerto Rico, Slovakia, Spain and Taiwan to receive efsitora once weekly or insulin degludec once daily administered subcutaneously. The primary objective of the study was to demonstrate non-inferiority in reducing A1C at week 26 with efsitora compared to insulin degludec. QWINT-4 (NCT05462756) was a parallel-design, open-label, treat-to-target, randomized controlled clinical trial comparing the efficacy and safety of efsitora as a weekly basal insulin to insulin glargine for 26 weeks in adults with type 2 diabetes who have previously been treated with basal insulin and at least two injections per day of mealtime insulin. The trial randomized 730 participants across the U.S., Argentina, Germany, India, Italy, Mexico, Puerto Rico and Spain to receive efsitora once weekly or insulin glargine once daily, both of which were administered subcutaneously along with insulin lispro. The primary objective of the trial was to demonstrate non-inferiority in reducing A1C at week 26 with efsitora compared to insulin glargine. About insulin efsitora alfa Insulin efsitora alfa (efsitora) is a once-weekly basal insulin, a fusion protein that combines a novel single-chain variant of insulin with a human IgG2 Fc domain. It is specifically designed for once-weekly subcutaneous administration, and with its low peak-to-trough ratio, it has the potential to provide more stable glucose levels (less glucose variability) throughout the week. About Lilly Lilly is a medicine company turning science into healing to make life better for people around the world. We've been pioneering life-changing discoveries for nearly 150 years, and today our medicines help tens of millions of people across the globe. Harnessing the power of biotechnology, chemistry and genetic medicine, our scientists are urgently advancing new discoveries to solve some of the world's most significant health challenges: redefining diabetes care; treating obesity and curtailing its most devastating long-term effects; advancing the fight against Alzheimer's disease; providing solutions to some of the most debilitating immune system disorders; and transforming the most difficult-to-treat cancers into manageable diseases. With each step toward a healthier world, we're motivated by one thing: making life better for millions more people. That includes delivering innovative clinical trials that reflect the diversity of our world and working to ensure our medicines are accessible and affordable. To learn more, visit and or follow us on Facebook, Instagram, and LinkedIn. P-LLY 1. The efficacy estimand represents the treatment effect on all participants who adhered to the study drug without initiating rescue therapy for persistent severe hyperglycemia. 2. From a baseline of 8.20% for efsitora and 8.28% for insulin glargine. 3. Participants treated with efsitora received a starting dose of 100 units of insulin, followed by escalation to fixed dosages of 150 units, 250 units and 400 units every 4 weeks, as needed, until achieving a target fasting blood glucose of 80-130 mg/dL. Participants with fasting blood glucose greater than 130 mg/dL on or after 16 weeks were transferred to flexible dosing. 4. From a baseline of 7.80% for both efsitora and insulin degludec. 5. From a baseline of 8.18% for both efsitora and insulin glargine. 6. The treatment-regimen estimand represents the estimated average treatment effect regardless of treatment discontinuation or introduction of rescue therapy for persistent severe hyperglycemia. 7. Blood glucose <54 mg/dL. 8. Nocturnal hypoglycemia was defined as any event that occurred at night between midnight and 6 a.m. Cautionary Statement Regarding Forward-Looking Statements This press release contains forward-looking statements (as that term is defined in the Private Securities Litigation Reform Act of 1995) about insulin efsitora alfa as a potential treatment for people with type 2 diabetes and the timeline for future readouts, presentations, and other milestones relating to insulin efsitora alfa and its clinical trials and reflects Lilly's current beliefs and expectations. However, as with any pharmaceutical product, there are substantial risks and uncertainties in the process of drug research, development, and commercialization. Among other things, there is no guarantee that future study results will be consistent with study results to date, that insulin efsitora alfa will prove to be a safe and effective treatment for type 2 diabetes, that insulin efsitora alfa will receive regulatory approval, or that Lilly will execute its strategy as expected. For further discussion of these and other risks and uncertainties that could cause actual results to differ from Lilly's expectations, see Lilly's Form 10-K and Form 10-Q filings with the United States Securities and Exchange Commission. Except as required by law, Lilly undertakes no duty to update forward-looking statements to reflect events after the date of this release. Trademarks and Trade Names All trademarks or trade names referred to in this press release are the property of the company, or, to the extent trademarks or trade names belonging to other companies are referenced in this press release, the property of their respective owners. Solely for convenience, the trademarks and trade names in this press release are referred to without the ® and ™ symbols, but such references should not be construed as any indicator that the company or, to the extent applicable, their respective owners will not assert, to the fullest extent under applicable law, the company's or their rights thereto. We do not intend the use or display of other companies' trademarks and trade names to imply a relationship with, or endorsement or sponsorship of us by, any other companies.
Arabian Business
17 hours ago
- Arabian Business
flydubai launches AI-powered biometric smart gates
Flydubai has introduced AI-driven smart gates for its crew members, aiming to streamline immigration procedures and enhance operational efficiency at Dubai airports. The UAE-based carrier has partnered with emaratech, a leading national technology provider, to deploy biometric border control systems at its Airport Operations Centre. The new smart gates leverage facial recognition, real-time data validation, and AI technology to create a seamless, paperless immigration experience for more than 3,800 pilots and cabin crew. Flydubai biometric smart gates This advanced digital-first solution significantly reduces immigration processing time, particularly during peak travel periods, and reflects the airline's ongoing commitment to innovation and punctuality. 'We are proud to partner with flydubai in pioneering a next-generation, paperless immigration experience,' said Thani Alzaffin, Group CEO of emaratech. 'Our biometric smart gates connect directly with both flydubai and immigration platforms for real-time identity validation, enabling a truly frictionless journey.' Thani Alzaffin, Group Chief Executive Officer of emaratech, said: 'We are proud to partner with flydubai in pioneering a next-generation, paperless immigration experience for their crew members. 'Through the integration of AI-powered facial recognition technology, our smart gates seamlessly connect with both flydubai's and immigration's platforms, enabling real-time validation and a truly frictionless journey.' The UAE-based carrier currently operates a modern fleet of 89 aircraft, serving more than135 destinations worldwide. The airline employs more than 6,400 staff, including over 1,300 pilots and 2,500 cabin crew. Mohammed Hareb AlMheiri, Chief Procurement and Technology Officer at flydubai, said: 'We always look for opportunities to harness the latest technologies that support our growth and operational efficiencies, and with the introduction of these biometric smart gates, this marks another step towards fostering a more seamless, punctual and secure operation.'
