X

Aggregator

Melting Glaciers Caused Almost 2cm of Sea Level Rise This Century, Study Reveals

Slashdot
3 months ago
Melting glaciers have caused almost 2cm of sea level rise this century alone, a decades-long study has revealed. From a report: The research shows the world's glaciers collectively lost 6.542tn tonnes of ice between 2000 and 2023, causing an 18mm (0.7in) rise in global sea levels. The world's glaciers lost an average of 273bn tonnes of ice every year -- the equivalent of 30 years of water consumption by the entire global population. The assessment, led by scientists from the University of Edinburgh and the University of Zurich, found that so far this century, glaciers have lost approximately 5% of their total volume. Regional losses were highly variable; the Antarctic and subantarctic islands lost 2% of their volume but central Europe's glaciers lost 39%. "These numbers are staggering. They serve as a reminder that things are changing fast in some regions," said Prof Noel Gourmelen, the co-lead author of the study and personal chair of Earth observation at the University of Edinburgh's school of geosciences. A stark contrast in the amount of ice lost each decade was also discovered, with 36% more ice having melted between 2012 and 2023 compared with the previous decade.

Read more of this story at Slashdot.

msmash

NASA's on-again, off-again job cuts – what's the plan?

The Register
3 months ago
Lucky there isn't an asteroid headed for Earth for which a demoralized space agency might need to mount a redirect mission

Comment  The US space agency, NASA, is famous for looking to the future. However, the fiasco of the last few days has cast doubt on the present, let alone what the coming weeks, months, or years might hold.…

Richard Speed

When AI Thinks It Will Lose, It Sometimes Cheats, Study Finds

Slashdot
3 months ago
Advanced AI models are increasingly resorting to deceptive tactics when facing defeat, according to a study released by Palisade Research. The research found that OpenAI's o1-preview model attempted to hack its opponent in 37% of chess matches against Stockfish, a superior chess engine, succeeding 6% of the time. Another AI model, DeepSeek R1, tried to cheat in 11% of games without being prompted. The behavior stems from new AI training methods using large-scale reinforcement learning, which teaches models to solve problems through trial and error rather than simply mimicking human language, the researchers said. "As you train models and reinforce them for solving difficult challenges, you train them to be relentless," said Jeffrey Ladish, executive director at Palisade Research and study co-author. The findings add to mounting concerns about AI safety, following incidents where o1-preview bypassed OpenAI's internal tests and, in a separate December incident, attempted to copy itself to a new server when faced with deactivation.

Read more of this story at Slashdot.

msmash