AI Insights
New York Passes RAISE Act—Artificial Intelligence Safety Rules
The New York legislature recently passed the Responsible AI Safety and Education Act (SB6953B) (“RAISE Act”). The bill awaits signature by New York Governor Kathy Hochul.
Applicability and Relevant Definitions
The RAISE Act applies to “large developers,” which is defined as a person that has trained at least one frontier model and has spent over $100 million in compute costs in aggregate in training frontier models.
- “Frontier model” means either (1) an artificial intelligence (AI) model trained using greater than 10°26 computational operations (e.g., integer or floating-point operations), the compute cost of which exceeds $100 million; or (2) an AI model produced by applying knowledge distillation to a frontier model, provided that the compute cost for such model produced by applying knowledge distillation exceeds $5 million.
- “Knowledge distillation” is defined as any supervised learning technique that uses a larger AI model or the output of a larger AI model to train a smaller AI model with similar or equivalent capabilities as the larger AI model.
The RAISE Act imposes the following obligations and restrictions on large developers:
- Prohibition on Frontier Models that Create Unreasonable Risk of Critical Harm: The RAISE Act prohibits large developers from deploying a frontier model if doing so would create an unreasonable risk of “critical harm.”
- “Critical harm” is defined as the death or serious injury of 100 or more people, or at least $1 billion in damage to rights in money or property, caused or materially enabled by a large developer’s use, storage, or release of a frontier model through (1) the creation or use of a chemical, biological, radiological or nuclear weapon; or (2) an AI model engaging in conduct that (i) acts with no meaningful human intervention and (ii) would, if committed by a human, constitute a crime under the New York Penal Code that requires intent, recklessness, or gross negligence, or the solicitation or aiding and abetting of such a crime.
- Pre-Deployment Documentation and Disclosures: Before deploying a frontier model, large developers must:
- (1) implement a written safety and security protocol;
- (2) retain an unredacted copy of the safety and security protocol, including records and dates of any updates or revisions, for as long as the frontier model is deployed plus five years;
- (3) conspicuously publish a redacted copy of the safety and security protocol and provide a copy of such redacted protocol to the New York Attorney General (“AG”) and the Division of Homeland Security and Emergency Services (“DHS”) (as well as grant the AG access to the unredacted protocol upon request);
- (4) record and retain for as long as the frontier model is deployed plus five years information on the specific tests and test results used in any assessment of the frontier model that provides sufficient detail for third parties to replicate the testing procedure; and
- (5) implement appropriate safeguards to prevent unreasonable risk of critical harm posed by the frontier model.
- Safety and Security Protocol Annual Review: A large developer must conduct an annual review of its safety and security protocol to account for any changes to the capabilities of its frontier models and industry best practices and make any necessary modifications to protocol. For material modifications, the large developer must conspicuously publish a copy of such protocol with appropriate redactions (as described above).
- Reporting Safety Incidents: A large developer must disclose each safety incident affecting a frontier model to the AG and DHS within 72 hours of the large developer learning of the safety incident or facts sufficient to establish a reasonable belief that a safety incident occurred.
- “Safety incident” is defined as a known incidence of critical harm or one of the following incidents that provides demonstrable evidence of an increased risk of critical harm: (1) a frontier model autonomously engaging in behavior other than at the request of a user; (2) theft, misappropriation, malicious use, inadvertent release, unauthorized access, or escape of the model weights of a frontier model; (3) the critical failure of any technical or administrative controls, including controls limiting the ability to modify a frontier model; or (4) unauthorized use of a frontier model. The disclosure must include (1) the date of the safety incident; (2) the reasons the incident qualifies as a safety incident; and (3) a short and plain statement describing the safety incident.
If enacted, the RAISE Act would take effect 90 days after being signed into law.
AI Insights
Tampa General Hospital, USF developing artificial intelligence to monitor NICU baby’s pain in real-time
TAMPA, Fla. – Researchers are looking to use artificial intelligence to detect when a baby is in pain.
The backstory:
A baby’s cry is enough to alert anyone that something’s wrong. But for some of the most critical babies in hospital care, they can’t cry when they are hurting.
READ: FDA approves first AI tool to predict breast cancer risk
“As a bedside nurse, it is very hard. You are trying to read from the signals from the baby,” said Marcia Kneusel, a clinical research nurse with TGH and USF Muma NICU.
With more than 20 years working in the neonatal intensive care unit, Kneusel said nurses read vital signs and rely on their experience to care for the infants.
“However, it really, it’s not as clearly defined as if you had a machine that could do that for you,” she said.
MORE: USF doctor enters final year of research to see if AI can detect vocal diseases
Big picture view:
That’s where a study by the University of South Florida comes in. USF is working with TGH to develop artificial intelligence to detect a baby’s pain in real-time.
“We’re going to have a camera system basically facing the infant. And the camera system will be able to look at the facial expression, body motion, and hear the crying sound, and also getting the vital signal,” said Yu Sun, a robotics and AI professor at USF.
Yu heads up research on USF’s AI study, and he said it’s part of a two-year $1.2 million National Institutes of Health grant.
He said the study will capture data by recording video of the babies before a procedure for a baseline. Video will record the babies for 72 hours after the procedure, then be loaded into a computer to create the AI program. It will help tell the computer how to use the same basic signals a nurse looks at to pinpoint pain.
READ: These states are spending the most on health insurance, study shows
“Then there’s alarm will be sent to the nurse, the nurse will come and check the situation, decide how to treat the pain,” said Sun.
What they’re saying:
Kneusel said there’s been a lot of change over the years in the NICU world with how medical professionals handle infant pain.
“There was a time period we just gave lots of meds, and then we realized that that wasn’t a good thing. And so we switched to as many non-pharmacological agents as we could, but then, you know, our baby’s in pain. So, I’ve seen a lot of change,” said Kneusel.
Why you should care:
Nurses like Kneusel said the study could change their care for the better.
“I’ve been in this world for a long time, and these babies are dear to me. You really don’t want to see them in pain, and you don’t want to do anything that isn’t in their best interest,” said Kneusel.
MORE: California woman gets married after lifesaving surgery to remove 40-pound tumor
USF said there are 120 babies participating in the study, not just at TGH but also at Stanford University Hospital in California and Inova Hospital in Virginia.
What’s next:
Sun said the study is in the first phase of gathering the technological data and developing the AI model. The next phase will be clinical trials for real world testing in hospital settings, and it would be through a $4 million NIH grant, Sun said.
The Source: The information used in this story was gathered by FOX13’s Briona Arradondo from the University of South Florida and Tampa General Hospital.
AI Insights
Ramp Debuts AI Agents Designed for Company Controllers
Financial operations platform Ramp has debuted its first artificial intelligence (AI) agents.
AI Insights
How automation is using the latest technology across various sectors
A majority of small businesses are using artificial intelligence and finding out it can save time and money.
Artificial Intelligence and automation are often used interchangeably. While the technologies are similar, the concepts are different. Automation is often used to reduce human labor for routine or predictable tasks, while A.I. simulates human intelligence that can eventually act independently.
“Artificial intelligence is a way of making workers more productive, and whether or not that enhanced productivity leads to more jobs or less jobs really depends on a field-by-field basis,” said senior advisor Gregory Allen with the Wadhwani A.I. center at the Center for Strategic and International Studies. “Past examples of automation, such as agriculture, in the 1920s, roughly one out of every three workers in America worked on a farm. And there was about 100 million Americans then. Fast forward to today, and we have a country of more than 300 million people, but less than 1% of Americans do their work on a farm.”
A similar trend happened throughout the manufacturing sector. At the end of the year 2000, there were more than 17 million manufacturing workers according to the U.S. Bureau of Labor statistics and the Federal Reserve Bank of St. Louis. As of June, there are 12.7 million workers. Research from the University of Chicago found, while automation had little effect on overall employment, robots did impact the manufacturing sector.
“Tractors made farmers vastly more productive, but that didn’t result in more farming jobs. It just resulted in much more productivity in agriculture,” Allen said.
ARTIFICIAL INTELLIGENCE DRIVES DEMAND FOR ELECTRIC GRID UPDATE
Researchers are able to analyze the performance of Major League Baseball pitchers by using A.I. algorithms and stadium camera systems. (University of Waterloo / Fox News)
According to our Fox News Polling, just 3% of voters expressed fear over A.I.’s threat to jobs when asked about their first reaction to the technology without a listed set of responses. Overall, 43% gave negative reviews while 26% reacted positively.
Robots now are being trained to work alongside humans. Some have been built to help with household chores, address worker shortages in certain sectors and even participate in robotic sporting events.
The most recent data from the International Federation of Robotics found more than 4 million robots working in factories around the world in 2023. 70% of new robots deployed that year, began work alongside humans in Asia. Many of those now incorporate artificial intelligence to enhance productivity.
“We’re seeing a labor shortage actually in many industries, automotive, transportation and so on, where the older generation is going into retirement. The middle generation is not interested in those tasks anymore and the younger generation for sure wants to do other things,” Arnaud Robert with Hexagon Robotics Division told Reuters.
Hexagon is developing a robot called AEON. The humanoid is built to work in live industrial settings and has an A.I. driven system with special intelligence. Its wheels help it move four times faster than humans typically walk. The bot can also go up steps while mapping its surroundings with 22 sensors.
ARTIFICIAL INTELLIGENCE FUELS BIG TECH PARTNERSHIPS WITH NUCLEAR ENERGY PRODUCERS
Researchers are able to create 3D models of pitchers, which athletes and trainers could study from multiple angles. (University of Waterloo)
“What you see with technology waves is that there is an adjustment that the economy has to make, but ultimately, it makes our economy more dynamic,” White House A.I. and Crypto Czar David Sacks said. “It increases the wealth of our economy and the size of our economy, and it ultimately improves productivity and wages.”
Driverless cars are also using A.I. to safely hit the road. Waymo uses detailed maps and real-time sensor data to determine its location at all times.
“The more they send these vehicles out with a bunch of sensors that are gathering data as they drive every additional mile, they’re creating more data for that training data set,” Allen said.
Even major league sports are using automation, and in some cases artificial intelligence. Researchers at the University of Waterloo in Canada are using A.I. algorithms and stadium camera systems to analyze Major League Baseball pitcher performance. The Baltimore Orioles joint-funded the project called Pitchernet, which could help improve form and prevent injuries. Using Hawk-Eye Innovations camera systems and smartphone video, researchers created 3D models of pitchers that athletes and trainers could study from multiple angles. Unlike most video, the models remove blurriness, giving a clearer view of the pitcher’s movements. Researchers are also exploring using the Pitchernet technology in batting and other sports like hockey and basketball.
ELON MUSK PREDICTS ROBOTS WILL OUTSHINE EVEN THE BEST SURGEONS WITHIN 5 YEARS
Overview of a PitcherNet System graphics analyzing a pitcher’s baseball throw. (University of Waterloo)
The same technology is also being used as part of testing for an Automated Ball-Strike System, or ABS. Triple-A minor league teams have been using the so-called robot umpires for the past few seasons. Teams tested both situations in which the technology called every pitch and when it was used as challenge system. Major League Baseball also began testing the challenge system in 13 of its spring training parks across Florida and Arizona this February and March.
Each team started a game with two challenges. The batter, pitcher and catcher were the only players who could contest a ball-strike call. Teams lost a challenge if the umpire’s original call was confirmed. The system allowed umpires to keep their jobs, while strike zone calls were slightly more accurate. According to MLB, just 2.6% of calls were challenged throughout spring training games that incorporated ABS. 52.2% of those challenges were overturned. Catchers had the highest success rate at 56%, followed by batters at 50% and pitchers at 41%.
GET FOX BUSINESS ON THE GO BY CLICKING HERE
Triple-A announced last summer it would shift to a full challenge system. MLB commissioner Rob Manfred said in June, MLB could incorporate the automated system into its regular season as soon as 2026. The Athletic reports, major league teams would use the same challenge system from spring training, with human umpires still making the majority of the calls.
Many companies across other sectors agree that machines should not go unsupervised.
“I think that we should always ensure that AI remains under human control,” Microsoft Vice Chair and President Brad Smith said. “One of first proposals we made early in 2023 was to insure that A.I., always has an off switch, that it has an emergency brake. Now that’s the way high-speed trains work. That’s the way the school buses, we put our children on, work. Let’s ensure that AI works this way as well.”
-
Funding & Business1 week ago
Kayak and Expedia race to build AI travel agents that turn social posts into itineraries
-
Jobs & Careers1 week ago
Mumbai-based Perplexity Alternative Has 60k+ Users Without Funding
-
Mergers & Acquisitions1 week ago
Donald Trump suggests US government review subsidies to Elon Musk’s companies
-
Funding & Business1 week ago
Rethinking Venture Capital’s Talent Pipeline
-
Jobs & Careers1 week ago
Why Agentic AI Isn’t Pure Hype (And What Skeptics Aren’t Seeing Yet)
-
Education3 days ago
9 AI Ethics Scenarios (and What School Librarians Would Do)
-
Education4 days ago
Teachers see online learning as critical for workforce readiness in 2025
-
Education1 week ago
AERDF highlights the latest PreK-12 discoveries and inventions
-
Education4 days ago
Nursery teachers to get £4,500 to work in disadvantaged areas
-
Education6 days ago
How ChatGPT is breaking higher education, explained