OpenAI CEO Sam Altman addressed concerns about AI's environmental impact by stating that human civilization itself has consumed vast amounts of energy.

Altman dismissed claims of AI consuming excessive water per query as "completely untrue, totally insane, no connection to reality," while acknowledging that data center cooling does require energy.

He argued that the responsibility for addressing AI's energy demands lies with the energy sector, which needs to transition rapidly towards nuclear or renewable sources. However, his remark that "It also takes a lot of energy to train a human" drew criticism online, with some calling it "dystopian" and "deeply antisocial and antihuman."

This debate also highlights the broader issue of transparency within the AI industry, as there are currently no regulations requiring data centers to disclose their energy and water consumption, and employees are often bound by non-disclosure agreements.

👉 Read the full article on Gizmodo

A significant bottleneck in quantum computing – the destruction of a qubit's state upon measurement – is showing signs of being overcome thanks to a new experimental approach.

Researchers have successfully demonstrated the ability to read the state of qubits designed to be more robust against environmental noise without destroying their quantum state, and in real-time. This addresses a long-standing challenge where observing a qubit often perturbs or collapses its state, hindering quantum computation.

The experiment focused on so-called "protected" qubits, particularly those related to topological qubits and Majorana modes, which theoretically distribute information to resist local noise. The breakthrough involved detecting the parity of such a system as a clear signal that changes between two states, indicating its quantum status without collapse.

While this does not signify the immediate arrival of fully functional quantum computers, it represents a crucial step in making qubit measurement a more manageable technical problem. This advancement paves the way for developing more stable quantum systems and ultimately, more powerful quantum computers.

👉 Read the full article on Gizmodo en Español