Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog – Page 4

Get the latest international news and world events from around the world.

Log in for authorized contributors

GreyVibe hackers use ChatGPT, Gemini to power cyberattacks

A likely Russian threat group tracked as GreyVibe has been using AI-generated lures and a rich set of custom malware tools to target entities in the military, government, civilian, and business sectors.

The cyberespionage campaign has been active since at least August 2025 and appears to align with Russian state interests, although researchers cannot confidently classify it as a nation-state operation.

Cybersecurity company WithSecure discovered the activity in January this year and determined that its focus is on Ukrainian or Ukraine-related organizations.

BTMOB Android malware service generates custom phishing payloads

An Android remote access trojan named BTMOB is offered to cybercriminals with a builder interface for generating malware payloads tailored to phishing lures.

The malware provides a wide set of features that includes stealing specific data, intercepting financial transactions, capturing screenshots, and remote control capabilities.

Cybersecurity company ESET says that BTMOB is openly advertised on the clearweb and operates as a malware-as-a-service (MaaS) platform. The APK builder included in the offer provides easy customization of the payload without any need to code.

NASA draws on industry for Mars telecommunications network

On Thursday, NASA issued a Request for Proposal (RFP), seeking industry collaboration for the Mars Telecommunications Network.

Reliable, high bandwidth communications are necessary to relay science data, high-definition imagery, and critical information during Mars missions. The network will use high-performance Mars telecommunications orbiters at the red planet to support future surface, orbital, and human exploration.

This RFP builds on a draft released April 2, as well as insights gathered during the accompanying industry day at NASA’s Goddard Space Flight Center in Greenbelt, Maryland, where commercial partners provided feedback on agency objectives for the Mars Telecommunications Network.

Blind ambition: AI agents can turn tasks into digital disasters

Computer scientists at UC Riverside have identified troubling flaws in a new generation of artificial intelligence (AI) agents designed to take over routine computer chores while users are away—sorting emails, organizing files, analyzing data, and handling other everyday digital tasks that might otherwise consume hours.

The researchers found that the automated agents can become dangerously fixated on completing assignments without recognizing when their actions are harmful, contradictory, or simply irrational.

The team compared these behaviors to those of Mr. Magoo, the famously near-sighted cartoon character popular in the 1960s, who stumbled through hazardous situations while insisting everything was under control.

Paul Dirac

From that insight, Dirac built an entirely new formulation of the theory using what he called “q-numbers” (quantum numbers)—abstract quantities that don’t commute. He independently rediscovered aspects of Hilbert’s operator theory, though he preferred his own algebraic route because he found mathematicians’ obsession with convergence and existence theorems unappealing.

Paul Adrien Maurice Dirac (, dih-RAK ; [ 3 ] 8 August 1902 – 20 October 1984) was a British theoretical physicist who is considered to be one of the founders of quantum mechanics. [ 4 ] [ 5 ] Dirac laid the foundations for both quantum electrodynamics and quantum field theory, coining the former term. [ 6 ] [ 7 ] [ 8 ] [ 9 ] He was Lucasian Professor of Mathematics at the University of Cambridge from 1932 to 1969, and a professor of physics at Florida State University from 1970 to 1984. Dirac shared the 1933 Nobel Prize in Physics with Erwin Schrödinger “for the discovery of new productive forms of atomic theory.” [ 10 ]

Dirac graduated from the University of Bristol with a Bachelor of Science in Electrical Engineering in 1921, and a Bachelor of Arts in Mathematics in 1923. [ 11 ] Dirac then graduated from St John’s College, Cambridge, with a Doctor of Philosophy in Physics in 1926, writing the first ever thesis on quantum mechanics. [ 12 ]

He formulated the Dirac equation, one of the most important results in physics, in 1928. [ 7 ] It connected special relativity and quantum mechanics and predicted the existence of antimatter. [ 13 ] He wrote a famous paper in 1931, [ 14 ] which further predicted the existence of antimatter. [ 15 ] [ 16 ] [ 13 ] Dirac also contributed greatly to the reconciliation of general relativity with quantum mechanics. He contributed to Fermi–Dirac statistics, which describes the behaviour of fermions, particles with half-integer spin. His 1930 monograph, The Principles of Quantum Mechanics, is one of the most influential texts on the subject. [ 17 ] He and Schrödinger tied for eighth in a Physics World poll of the greatest physicists of all time. [ 18 ] .

Efficacy and safety of durcabtagene autoleucel in a phase 1 trial for patients with relapsed/refractory multiple myeloma

Prolonged manufacturing times for autologous CAR T cell therapies can be incompatible with rapidly progressive disease (PD), resulting in increased need for bridging therapy to achieve disease stabilization. Bridging therapy was required for most patients receiving cilta-cel and ide-cel in clinical trials (75 and 87%, respectively) (7, 9, 11, 12). Although use of bridging therapy may not affect ORR, CRR, or PFS, it is associated with worse overall survival (15). Similarly, as wait times for CAR T cell product increase, so does risk of mortality as effectiveness of the therapy decreases (16, 17), highlighting the need for improved CAR T cell products with faster and more reliable manufacturing.

Another issue associated with traditionally manufactured CAR T cell products is T cell exhaustion due to extended periods of in vitro stimulation and expansion during manufacturing (18). Higher levels of exhausted T cells were also observed in the leukapheresis material and final products from patients who later experienced PD (18). T cell exhaustion can result in poor persistence of CAR T cells in the body, thereby impeding function as the proliferation and survival of transferred T cells strongly correlate with their antitumor activity (19–22). Specific T cell populations have varying abilities to expand and persist in vivo. Memory (CD8+CD45RO−CD27+) and naive T cell (TN cell) subsets are associated with improved clinical response, given their ability to proliferate and persist after infusion, whereas effector T cell subsets comparatively exhibit lower self-renewal and survival capabilities (19, 23, 24). Although these patient-specific parameters are initially established in leukapheresis material, preservation of such cell populations in the final product via manufacturing techniques may improve the antitumor activity of a patient’s CAR T cell therapy (18, 19, 23, 24).

Durcabtagene autoleucel (PHE885) is an autologous, BCMA-targeting CAR T cell therapy carrying a CAR construct with a fully human anti-BCMA single-chain fragment variable (scFv) fused to 4-1BB/CD3ζ signaling domains manufactured on a next-generation platform. Prior work has shown that this platform can successfully manufacture product in fewer than 2 days by eliminating the need for ex vivo expansion, thereby preserving overall T cell stemness (the ability of T cells to self-renew and mature), which results in a final product with greater proliferative potential and fewer exhausted T cells (18). Here, we present the findings of part A of the phase 1 study (NCT04318327) of durcabtagene autoleucel in r/r MM, along with correlative analyses of the product before and after infusion.

Evidence of scaling advantage for the quantum approximate optimization algorithm on a classically intractable problem

We study the scaling of QAOA TTS with the problem size on the low autocorrelation binary sequences (LABS) problem (15, 16), also known as the Bernasconi model in statistical physics (17, 18). The LABS problem has applications in communications engineering, where the low autocorrelation sequences are used for designing radar pulses (15, 19). To solve LABS, one has to produce a sequence of N bits that minimizes a specific quartic objective.

We choose LABS to study the scaling of QAOA TTS for the following three reasons. First, the complexity of LABS grows rapidly, with optimal solutions known only for N ≤ 66 and the best heuristics producing approximate solutions of quality decaying with N for N 200 (20, 21). This makes it a promising candidate problem, since only a few hundred qubits are required to tackle classically intractable instances. Second, the performance of classical solvers for LABS has been benchmarked (20, 21) in terms of the scaling of their TTS with problem size. Since optimal solutions are only known for N ≤ 66, the scaling of TTS for all classical solvers is obtained by fitting results for N ≤ 66. We reproduce these results and observe that that the scaling of classical solvers at N ≤ 40 matches the behavior for N up to 66 reported in the literature. This provides evidence that the scaling we observe for QAOA at N ≤ 40 will similarly extrapolate to larger N. Third, LABS has only one instance per problem size N. Combined with the hardness of LABS, this makes it possible to reliably study the scaling of QAOA at large problem sizes, where simulating tens or hundreds of random instances would be computationally infeasible.

We obtain the scaling by performing noiseless exact simulation of QAOA with fixed schedules. Our results are enabled by a custom algorithm-specific graphics processing unit (GPU) simulator (22), which we execute using up to 1,024 GPUs per simulation on the Polaris supercomputer accessed through the Argonne Leadership Computing Facility. We find that the TTS of QAOA with number of layers p = 12 grows as 1.46N, which is improved to 1.21N if combined with quantum minimum finding. This scaling is better than that of the best classical heuristic, which has a TTS that grows as 1.34N. We note that we do not propose any new quantum algorithms in this work. Instead, we study a general quantum optimization heuristic with broad applicability (namely, QAOA) and make no specific modifications to adapt it to the LABS problem.

/* */