Newsletter
Highlights:
C3.ai sees consistent, strong revenue growth in their Q4 results.
Meta has identified AI-generated content on political posts on their platform.
Palantir lands a $480M Army contract despite dropping stock values.
Business
C3.ai Reports Strong Q4 Results, Sees ‘Huge Opportunity’ in GenAI (PYMNTS)
C3.ai, a software company focused on enterprise AI applications, reported strong fourth-quarter results for the fiscal year 2024. This is their fifth consecutive quarter of accelerating revenue growth, exceeding analysts’ predictions. One of the things that sets C3.ai apart from other companies, according to CEO Tom Siebel, is their focus on addressing issues such as hallucinations, IP liability, and access control in generative AI.
Google’s generative AI fails 'will slowly erode our trust in Google’ (Yahoo!Finance)
Google’s AI Overview is supposed to use generative AI to answer your questions faster by placing the AI answer at the top of your search. However, it’s been giving people horrifically false results like putting glue on pizza and eating rocks for dietary health. Their temporary solution is to manually remove the worst results from the billions of queries. Chinmay Hegde, associate professor of computer science and engineering at NYU, explains that “Google is supposed to be the premiere source of information on the internet, and if that product is watered down, it will slowly erode our trust in Google.”
Meta identifies networks pushing deceptive content likely generated by AI (Reuters)
Facebook and Instagram have recently seen an influx of AI-generated content and comments on political posts that praise Israel’s handling of the war in Gaza. Meta recently identified the culprit as Tel Aviv-based political marketing firm STOIC. Meta’s threat investigations team is working diligently, identifying in a report six total covert influence operations that they’ve shut down in the first quarter. With elections in the EU in June and the USA in November, Meta will be putting their defenses to the test.
Government
Palantir lands $480M Army contract for Maven artificial intelligence tech (DefenseScoop)
Palantir has been awared a $480 million deal for its Maven Smart System prototype by the Army. According to the release sent out by the Pentagon, “The MAVEN Smart System (MSS) by Palantir along with National Geospatial Agency (NGA) Broad Area Search – Targeting (BAS-T) uses AI generated algorithms and memory learning capabilities to scan and identify enemy systems in the Area of Responsibility (AOR).” Defense officials hope to leverage AI to make faster and better decisions and improve operational effectiveness and efficiency.
Technology
Robots play soccer at Geneva AI showcase (Reuters)
At the AI for Good Global Summit in Geneva, teams of robots shambled around on a miniature soccer pitch, using input from sensors to kick, pass, and keep track of the ball. The Summit also featured prosthetic limbs that could learn from a user’s behavior, devices for visually impaired people to avoid obstacles in the street, and bionic companion pets. "Sometimes we think about AI as just something big," said Tomas Lamanauskas, Deputy Secretary-General of the U.N.'s International Telecommunication Union (ITU), "At the same time AI can be embedded in so many more things in everyday life...”
AI headphones let wearer listen to a single person in a crowd, by looking at them just once (Univ of Washington)
A University of Washington team has developed an artificial intelligence system that lets a user wearing headphones isolate sounds in their environment. The user just needs to look at a person speaking for three to five seconds and press the button on the headset. The system, called “Target Speech Hearing,” then identifies the voice and cancels all other sounds in the environment, even as the listener moves around in noisy places or no longer faces the speaker. The team is working to expand the system to earbuds and hearing aids in the future.
Life
How it feels to get an AI email from a friend (Neven Mrgan)
When Neven Mrgan received a letter from a friend, he didn’t know how to feel reading the disclosure at the bottom that it was written by AI. It wasn’t a test of the technology or a funny joke between friends, but simply that the friend had something to ask and for reasons unknown to Mrgan used AI to do it. Mrgan explores how this made him feel, comparing it to as if their friend “had buzzed the secretary over the intercom and barked at them to send [him] a letter” or “decorating a family fridge with printed stock art”. Even if the technology advances to capture the tone of his friend convincingly, Mrgan laments the possibility that human interaction will be outsourced to servers.
