Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog – Page 2215

Only a year ago, ChatGPT woke the world up to the power of foundation models. But this power is not about shiny, jaw-dropping demos. Foundation models will permeate every sector, every aspect of our lives, in much the same way that computing and the Internet transformed society in previous generations. Given the extent of this projected impact, we must ask not only what AI can do, but also how it is built. How is it governed? Who decides?

We don’t really know. This is because transparency in AI is on the decline. For much of the 2010s, openness was the default orientation: Researchers published papers, code, and datasets. In the last three years, transparency has waned. Very little is known publicly about the most advanced models (such as GPT-4, Gemini, and Claude): What data was used to train them? Who created this data and what were the labor practices? What values are these models aligned to? How are these models being used in practice? Without transparency, there is no accountability, and we have witnessed the problems that arise from the lack of transparency in previous generations of technologies such as social media.

To make assessments of transparency rigorous, the Center for Research on Foundation Models introduced the Foundation Model Transparency Index, which characterizes the transparency of foundation model developers. The good news is that many aspects of transparency (e.g., having proper documentation) are achievable and aligned with the incentives of companies. In 2024, maybe we can start to reverse the trend.

Read more | >

Boeing and SpaceX have shared footage of U.S. Space Force’s secretive X-37B mini-shuttles in space with a payload-laden service module attached. A brief video clip showing the X-37B with the module separating from its launch rocket after being lofted into space in 2020 was included in a video montage shown ahead of the latest launch of an X-37B yesterday. You can find out more about what we can expect from the new X-37B mission in The War Zone’s previous reporting.

SpaceX broadcast the video montage that included the clip in question just minutes before a Falcon Heavy rocket with an X-37B on top blasted off from the Kennedy Space Center in Florida last night. User @DutchSpace on X, formerly known as Twitter, was among the first to spot the clip of the X-37B separating into space.

The montage begins at approximately 3:38 in the runtime of the video seen below.

Read more | >

Fourier Intelligence has been manufacturing exoskeletons and rehabilitation devices since 2017. The Singapore-based company launched its first generation of humanoid robots this year, designated the GR-1.

The humanoid platform includes 40 degrees of freedom distributed throughout its body, which measures 1.65 m (5 ft., 5 in.) in height and weighs 55 kg (121.2 lb.). The joint module that is fitted at the hip of the robot is capable of producing a peak torque of 300 Nm, which allows it to walk at a speed of 5 kph (3.1 mph) and carry goods that weigh 50 kg (110.2 lb.).

Making the leap from exoskeleton development to humanoid design is a logical progression, as the humanoid platform shares many of the mechanical and electrical design elements that Fourier developed for its core product line. Actuation is a core competency of the company, and by designing and building actuators, it claimed that it can optimize the cost/performance of the system.

Read more | >

For decades, a substantial number of proteins, vital for treating various diseases, have remained elusive to oral drug therapy. Traditional small molecules often struggle to bind to proteins with flat surfaces or require specificity for particular protein homologs. Typically, larger biologics that can target these proteins demand injection, limiting patient convenience and accessibility.

In a new study published in Nature Chemical Biology, scientists from the laboratory of Professor Christian Heinis at EPFL have achieved a significant milestone in drug development. Their research opens the door to a new class of orally available drugs, addressing a long-standing challenge in the pharmaceutical industry.

“There are many diseases for which the targets were identified but drugs binding and reaching them could not be developed,” says Heinis. “Most of them are types of cancer, and many targets in these cancers are protein-protein interactions that are important for the tumor growth but cannot be inhibited.”

Read more | >

1. AGI could be achieved or we will get even closer. There will OpenAI releasing GPT5 and updates of Google LLM like improved Gemini.

Definition’s for AI AGI = artificial general intelligence = a machine that performs at the level of an average (median) human.

ASI = artificial superintelligence = a machine that performs at the level of an expert human in practically any field.

Read more | >

NASA has pushed forward a revolutionary new rocket technology at its Marshall Space Flight Center in Huntsville, Alabama. Engineers at the facility fired the 3D-printed Rotating Detonation Rocket Engine (RDRE) for a record 251 seconds with 5,800 lb (2,631 kg) of thrust.

For over six decades, NASA has relied on chemical rockets to launch its vehicles into space. It works, but chemical rockets suffer from the fact that they’ve been operating in the neighborhood of their theoretical limit since 1942. This isn’t helped by the fact that most liquid rockets are essentially unchanged in their basic design since the days of the German V2s.

To squeeze a bit more performance out of rocket engines, NASA is looking at a fundamentally different design with the RDRE.

Read more | >

ChatGPT may do an impressive job at correctly answering complex questions, but a new study suggests it may be absurdly easy to convince the AI chatbot that it’s in the wrong.

A team at Ohio State University challenged (LLMs) like ChatGPT to a variety of debate-like conversations in which a user pushed back when the chatbot presented a correct answer.

Through experimenting with a broad range of reasoning puzzles, including math, common sense, and logic, the study found that when presented with a challenge, the model was often unable to defend its correct beliefs and instead blindly believed invalid arguments made by the user.

Read more | >