Ai
OpenAI CEO and AI Critic Clash Over Hype and Reality
A public spat between OpenAI CEO Sam Altman and cognitive scientist Gary Marcus highlights a growing rift over AI's promises versus its actual capabilities. Marcus accused Altman of overhyping AI, while Altman defended his company's achievements, underscoring a broader industry tension.
OpenAI Warns of Fraudulent Equity Schemes Targeting AI Investors
OpenAI has issued a public warning about unauthorized offers to buy its equity, cautioning investors against fraudulent schemes that could result in financial loss and legal liability. The company emphasizes that any transfer without its written consent is void and may violate securities laws.
Purdue Study: ChatGPT Gives Wrong Coding Answers More Than Half the Time
A Purdue University study presented at a human-computer interaction conference reveals that 52% of ChatGPT's answers to programming questions contain misinformation, and 77% are more verbose than human responses. The research, based on 517 Stack Overflow questions, highlights that programmers often prefer ChatGPT's polite tone, which can mask errors.
CIA Builds AI Chatbots to Simulate Foreign Leaders' Reactions
The CIA has developed a platform using large language models to let analysts interact with simulated versions of foreign leaders, aiming to predict their decisions. The initiative, led by the agency's first CTO, seeks to modernize intelligence work and streamline private-sector partnerships, though details remain classified.
Google’s Bard Chatbot Backs DOJ in Antitrust Case, Calls for Breakup
Days after its public release, Google's AI chatbot Bard sided with the Justice Department in an antitrust exchange, arguing the company holds a monopoly in digital ads. The bot's response, shared by a security researcher, highlights ongoing federal lawsuits against Google. This episode underscores the tension between AI tools and their corporate creators.
DeepSeek's Next AI Model Poised to Test US Tech Market Again
DeepSeek is set to release its V4 AI model, a move that could rattle US tech markets already strained by massive AI investments. The launch follows the V3 debut that caused a 3% Nasdaq drop and a $600 billion Nvidia loss. Investors and tech giants are bracing for potential market volatility.
Bezos and Musk Beam Into Star Trek Pilot via Deepfake
A new deepfake video places Amazon's Jeff Bezos and SpaceX's Elon Musk into the original 1965 Star Trek pilot, with Bezos appearing as a Talosian alien. The clip highlights both the creative potential and the serious ethical concerns surrounding AI-generated likeness manipulation, including nonconsensual pornography and disinformation risks.
Writer Stands by AI Assistance After Book's False Quotes Exposed
Steven Rosenbaum, author of a book on AI and truth, faces backlash after The New York Times found fabricated quotes in his work, which he attributes to AI hallucinations. Despite the scandal, Rosenbaum defends his use of AI tools, calling them a 'delightful writing companion,' and says he won't return to traditional writing methods.
Lyft Rolls Out Autonomous Rides in Bay Area via Drive.ai Partnership
Lyft has begun offering self-driving car rides to select passengers in the San Francisco Bay Area, using technology from startup Drive.ai. The move aligns with Lyft's broader vision of a future dominated by autonomous fleets and a decline in personal car ownership, as outlined by co-founder John Zimmer.
Music Labels and Google Explore Licensing AI-Generated Vocals
Universal Music Group, after previously demanding removal of AI-cloned songs, is reportedly negotiating with Google to license artists' voices and melodies for fan-generated tracks. Warner Music is also involved, with executives suggesting such tools could offer new creative avenues while artists may opt out. The talks signal a potential shift toward monetizing AI deepfakes in the music industry.