The first in-womb treatment for motor-neuron condition is a success
Creating a world that responds to player inputs from online multiplayer game Bleeding Edge with a World and Human Action Model
There are severe consequences of withdrawing or drastically reducing long-standing commitments on the international front. The United States is often the largest contributor to global initiatives directly linked to the United Nations Sustainable Development Goals (SDGs), the world’s plan to end poverty and achieve environmental sustainability. One-fifth of the core budget of the World Health Organization is provided by the country, despite the fact Donald Trump has decided to leave. The WHO has told all its staff to put their travel on hold, because the United States won’t formally leave until next year.
A new artificial intelligence engine is able to create a video-game world that responds to player inputs. Researchers trained the engine — the World and Human Action Model (WHAM) — on one-second slices of gameplay from the online multiplayer game Bleeding Edge, taken from controller inputs from 500,000 anonymized play sessions. They showed that WHAM’s creations adhere to the original game rules, and can generate a variety of game scenarios that retain items added by the player.
Microsoft Has Done It All: The First Topological Quantum Computational Bits for a Baby with Motor-Neuron Disorder
A two-and-a-half-year-old girl shows no signs of a rare genetic disorder, after becoming the first person to be treated for the motor-neuron condition while in the womb. The mother took the drug when she was pregnant and the child is still taking it. The drug, Risdiplam, modifies the expression of a gene to help maintain motor cells in the brain and spine.
Microsoft hopes to underpin a future generation of quantum computers by creating the first topological qubits, a way of storing quantum information. The announcement came in a press release containing few technical details, which has left some researchers sceptical about the company’s claims. Microsoft said that it has handed over data to specialists. “Would I bet my life that they’re seeing what they think they’re seeing? No, but it looks pretty good,” says theoretical physicist Steven Simon, who was among those briefed.
The Chinese AI ecosystem is growing too fast. Science and science needs to change, but scientists need to learn how to make the most out of China
It is hard to put into words the extent of the damage being done to the US research enterprise, which is of almost incalculable value to both the nation itself and the wider world. Organizations representing the global scientific communities are beginning to react. They need to speak out for their US colleagues. The US educational and scientific communities need to know they are not alone. An assault on science and scientists is an assault on science and scientists everywhere.
Experts warn there might be more disruption to come. The proposals put forth in Project 2025, a blueprint organized by the Heritage Foundation, have been rolled out so far. The document also called for slashes to climate research and new rules that could make it easier to fire government employees, including scientists, who were hired to positions on the basis of expertise rather than being politically appointed.
China has been trying to get away from the image that their nation lags behind the US in innovation for a decade. The launch of the large-language model DeepSeek-R1 has now put the spotlight on China’s wider artificial intelligence (AI) ecosystem. Chinese artificial intelligence companies are more focused on solving industrial and manufacturing problems at scale than designing consumer-facing products. The victor of the race might become less relevant as the two countries have different priorities.
Source: Daily briefing: First in-womb treatment for motor-neuron condition is a success
Why science is important, why science matters: How scientists and the public can work together to improve the fight against rat life in the 21st century
cities around the world are fighting rats because of a warming climate. Efforts to control urban rat populations rely heavily on poisons and traps, cruel and largely ineffective measures that can have knock-on effects on other wildlife. Thankfully, research has shown that there’s another way (and bonus, it’s not too difficult to implement): we keep our rubbish better contained. Some careful planning is needed, but some examples such as New York City’s Containerization approach show that it works. All without the need for a gruesome end to rat life.
Normally cells have cellular machinery that removes non-coding sequences from immature transcripts and replaces them with coding sequences called exons. The formation of an amino-acid chain is aided by the creation of the mRNA sequence. In a tumour (b), splicing can go wrong. In those instances, the cells produce incorrect or unusual proteins. The proteins produce structures called neoantigens, which flag the cells to the immune system as ‘foreign’. 7 min read, Nature paywall.
In an editorial, Nature Computational Science commits to fostering a culture of diversity and inclusion among their authors, readers and communities. (Nature Computational Science | 4 min read)
Those who represent young scientists, scientific Academies and researchers at risk around the world need to be supported by global scientific organizations. We urge them all to speak up for their US-based colleagues — and the crucial work they do — just as they support researchers at risk elsewhere.
There are differences of opinion between researchers in all scientific fields. Discuss and study is the best way to understand each other. Shutting down scholarship is not a solution.
Attacks on people’s rights and academic freedom are unacceptable. They will halt, if not reverse, decades of progress in scientific research. Worldwide, the research consensus is that more granular, more disaggregated data are essential to achieving the SDGs. We also know that incorporating sex and gender into study design improves science and medicine, including diagnosis and treatment.
Another executive order mandated the cancellation of policies and initiatives in diversity, equity and inclusion, which the Trump administration has incorrectly described as illegal, radical and wasteful. Cancelling public initiatives that create inclusive and welcoming spaces, and stopping federal funding for the study of these issues, could cause harm. Equally, there is no justification for cancelling policies and funding for initiatives that recognize the complexity of sex and gender, something the White House wrongly calls “gender ideology extremism”.
Federal agencies and universities are affected by turmoil. Thousands of researchers are in limbo as they await word on when the funding freeze will thaw. Millions of grant recipients around the world have been abandoned by the US.
One million women in countries around the world have lost access to contraceptive care because of a 90-day halt on funding from the US Agency for International Development. In the year of 2020, the United States provided over seventy billion dollars in international assistance.
Trump has also cancelled US federal funding for international climate-change projects, which totalled some US$11 billion in 2024, amounting to around 10% of annual global public climate finance. Alongside his decision to withdraw from the 2015 Paris climate agreement, this is a severe blow to tackling climate change, and will delay efforts to boost finance for the countries most affected by global warming.
Individuals and organizations are challenging some of the Trump administration’s actions in the courts, and it might be that the White House is forced to moderate or reverse some of its decisions. The direction of travel is clear, as there is a desire to change if not remove, independent, science based evidence and expert advice.
Researchers trained an artificial intelligence engine on one-second slices of gameplay from online multiplayer game ‘B Bleeding Edge’, taken from controllers’ inputs. The engine, World and Human Action Model (WHAM), can create a world that responds to player inputs. WHAM was trained on one-second slices of gameplay taken from 5,00,000 anonymized play sessions.