The FAA launched a new AI system called SMART this week, deploying it first in the Washington, D.C. region at Reagan National, Dulles International, and BWI Marshall airports. The system pulls from 200 data streams including weather forecasts, airline schedules, and flight paths to predict delays before they happen. Transportation Secretary Sean Duffy said SMART gives controllers better real-time information, while FAA Administrator Bryan Bedford noted it helped divert 11 flights during a recent storm.
SMART is part of the Flow Management Data and Services program and represents a major investment. The FAA awarded Air Space Intelligence a 12-year, $875 million contract in June to build the system. Officials emphasized that air traffic controllers will always retain final responsibility for separating aircraft, responding to critics who question using AI in safety-critical roles.
In other tech news, Google announced its first Googlebook AI laptops starting at $899, developed in partnership with Acer, Asus, Dell, HP, and Lenovo. The machines feature 13 to 15 inch touch screens, up to 14 hours of battery life, and built-in Gemini AI capabilities. They will run on Intel Core Ultra Series 3 or Qualcomm Snapdragon X chips, with pre-orders opening September 21 and US shipments starting October 4.
A federal report warns that the US power grid needs significant upgrades to handle growing AI data center demand. The grid requires thousands of miles of new power lines and billions in investment, with about 4,000 miles needing replacement annually for decades. Electricity demand is projected to rise 16 to 25 percent by 2034. Officials stressed that data center operators should cover upgrade costs, not everyday families.
The global chip market is on track to surpass $1.6 trillion this year, driven by AI infrastructure spending. Samsung Electronics and SK Hynix are leading growth in memory chip sales, with DRAM and NAND flash prices climbing sharply. Chip industry employment hit 174,000 in the first quarter, up 15 percent from 2018, and the market is expected to approach $2 trillion by 2030.
Key Takeaways
- The FAA launched AI system SMART at D.C.-area airports to predict flight delays using 200 data streams
- FAA awarded Air Space Intelligence a 12-year, $875 million contract to develop SMART and the FMDS platform
- Googlebook AI laptops start at $899 with Gemini features, Intel Core Ultra or Snapdragon X chips, shipping October 4
- US power grid needs thousands of miles of new lines and billions in investment for AI data center demand
- Electricity demand expected to rise 16 to 25 percent by 2034 due to AI data centers
- Global chip market forecast to top $1.6 trillion in 2026, driven by AI infrastructure investment
- Samsung Electronics and SK Hynix lead memory chip growth as DRAM and NAND prices rise sharply
- Chip industry employment reached 174,000 in Q1, up 15 percent from 2018
- SMART uses color-coded displays to show how long planes have been waiting at Reagan National, Dulles, and BWI
- Officials stress humans retain final authority over airspace decisions despite AI assistance
FAA launches AI system SMART to predict flight delays
Transportation Secretary Sean Duffy and FAA Administrator Bryan Bedford unveiled the FAA's new AI system called SMART on Monday. The system combines 200 data streams including weather forecasts, airline schedules, and flight paths to predict delays before they happen. FAA officials say SMART could reduce delays, cancellations, and fuel use while easing controller workload. The system is part of the Flow Management Data and Services program meant to replace the older Traffic Flow Management System.
FAA deploys SMART AI tool for air traffic in DC region
The FAA and Department of Transportation launched the AI tool called SMART in the Washington, D.C. area on Monday. Transportation Secretary Sean Duffy said SMART gives controllers better real-time data about current and future flight conditions. The tool shows plane status on screen using color changes to indicate how long a plane has been waiting. SMART is now active at Reagan National, Dulles International, and BWI Marshall airports and will expand to other regions later.
FAA launches AI system to predict flight delays
The FAA announced a new cloud-based AI system called SMART on Monday that combines airline schedules, weather data, and airport capacity to predict traffic flows. The system gives the FAA and airlines a shared view before departure so they can plan more efficient routes and times. The FAA awarded Air Space Intelligence a 12-year, $875 million contract in June to develop SMART and the related FMDS platform. Officials stressed the technology is meant to support air traffic controllers who retain final responsibility for separating aircraft.
FAA uses AI technology SMART to cut flight delays in DC
The FAA launched new AI technology called SMART on Monday in the D.C. region to reduce flight delays and cancellations. FAA Administrator Bryan Bedford said the system helped divert 11 flights during a recent storm by using better weather data. The technology is now active at Reagan National, BWI Marshall, and Dulles airports. Some critics including Congressman Don Beyer expressed concern about using unproven AI in air traffic control. Transportation Secretary Sean Duffy stressed that humans will always control airspace and make decisions.
Google launches Googlebook AI laptops starting at $899
Google announced pricing for its first Googlebook AI laptops on September 21, with prices starting at $899. The laptops are being made with five manufacturers including Acer, Asus, Dell, HP, and Lenovo. They feature 13 to 15 inch touch screens, up to 14 hours of battery life, and built-in Gemini AI features. The laptops will be powered by Intel Core Ultra Series 3 or Qualcomm Snapdragon X chips and will ship in the US starting October 4.
Google launches premium Googlebook AI laptops with partners
Google announced its first Googlebook AI laptops priced at $899 and up. The laptops are made with Acer, Asustek, Dell, HP and Lenovo. They feature 13 to 15 inch touch screens and up to 14 hours of battery life. The machines use Intel Core Ultra or Qualcomm Snapdragon X chips. Pre orders opened on Sept 21 with US shipments starting Oct 4.
Physical AI could transform how factories collect operational data
Manufacturing operations currently rely on manual systems that record what was said to happen rather than what actually happened. This data gap limits industrial AI because models cannot learn well from incomplete records. The article calls for systems of observation using sensors and cameras to capture real operations. This shift could improve root cause analysis and worker empowerment over a long transformation.
Power grid needs major upgrades for AI data center demand
A new federal report says the US power grid needs thousands of miles of new power lines and billions in investment. About 4,000 miles of lines will need replacement each year for decades. Most transmission infrastructure was built in the 1960s and 1970s. Electricity demand is expected to rise 16 to 25 percent by 2034 due to AI data centers. Officials said data center operators should pay for upgrades, not families.
Global chip market expected to top $1.6 trillion in 2026
The global semiconductor market is forecast to surpass $1.6 trillion this year driven by AI infrastructure investment. Samsung Electronics and SK Hynix lead growth with memory chip sales. DRAM and NAND flash prices have risen sharply compared with a year earlier. Chip industry employment reached 174,000 in the first quarter, up 15 percent from 2018. The market is expected to reach about $2 trillion by 2030.
Washington University launches AI teaching toolkit for faculty
Washington University in St. Louis started the AI Curriculum Corps to help faculty learn about AI. The program encourages thoughtful and ethical use of AI tools in teaching. Anna Cunningham said good teaching methods should guide AI use, not replace it. Faculty explored how AI can support learning goals and rethink assessments. The program values diverse perspectives on AI in education.
KT joins global AI security standards group at ITU-T
KT Corp., South Korea's largest telecom operator, was selected as a member of the International Telecommunication Union Study Group 17. The group focuses on developing global standards for AI security and digital trust. KT was chosen for its expertise in these areas. The company aims to help ensure AI systems are secure and trustworthy worldwide.
AI systems in finance are too connected to fail safely
A new analysis from the University of California, Berkeley finds that AI in the financial sector is highly interconnected. If one AI system fails, it could cause widespread disruptions across the industry. Researchers used network analysis to map connections between financial institutions. The study calls for stronger resilience and more proactive monitoring by regulators.
Runaway AI development does not serve China's interests
An opinion piece discusses China's Minister of State Security Chen Yixin's recent article on AI governance. The piece argues that uncontrolled AI development poses risks China does not want. Chen highlighted the need for stronger legal safeguards, including updates to the Cybersecurity Law and Counter-Espionage Law. AI developers in China may face serious consequences if their systems cause harm.
Wharton and Bank of England discuss AI finance rules
Professor Itay Goldstein of the Wharton School spoke with Sarah Breeden, deputy governor for financial stability at the Bank of England. They discussed how regulators can manage risks from agentic AI in financial markets. The conversation covered human oversight, risk mitigation tools, and the balance between innovation and stability. This episode is part of the Future of Finance series.
Sources
- FAA to launch AI system to predict flight delays before takeoff
- FAA rolls out air traffic AI tool in Washington region
- FAA to launch AI system to predict flight delays before takeoff
- FAA now using artificial intelligence in effort to reduce flight delays in the DC area
- Google rolls out US$899-plus ‘Googlebook’ AI laptops with hardware partners
- Google rolls out $1,150-plus ‘Googlebook’ AI laptops with hardware partners
- Seeing Is Believing: How Operational Insights Are Advancing With Physical AI
- Power Grid Faces Costly Upgrades Due To AI Data Centers
- Global chip market projected to exceed $1.6 trillion this year despite AI slowdown
- A new toolkit for teaching in the age of AI
- KT drives global standards for AI security and digital trust at ITU-T SG17
- AI is becoming too interconnected to fail
- Thought for the week: Why runaway AI development does not serve China's interests
- AI and Financial Regulation
Comments
Please log in to post a comment.