arXiv subjects

David Duncan

Publications and source records attributed to David Duncan.

3 recordsLinked to original sources

arXiv2025

Beyond Automation: Redesigning Jobs with LLMs to Enhance Productivity

The adoption of generative artificial intelligence (AI) is predicted to lead to fundamental shifts in the labour market, resulting in displacement or augmentation of AI-exposed roles. To investigate the impact of AI across a large organisation, we assessed AI exposure at the task level within roles at the UK Civil Service (UKCS). Using a novel dataset of UKCS job adverts, covering 193,497 vacancies over 6 years, our large language model (LLM)-driven analysis estimated AI exposure scores of 1,542,411 tasks. By aggregating AI exposure scores for tasks within each role, we calculated the mean and variance of job-level exposure to AI, highlighting the heterogeneous impacts of AI, even for seemingly identical jobs. We then use an LLM to redesign jobs, focusing on task automation, task optimisation, and task reallocation. We find that the redesign process leads to tasks where humans have comparative advantage over AI, including strategic leadership, complex problem resolution, and stakeholder management. Overall, automation and augmentation are expected to have nuanced effects across all levels of the organisational hierarchy. Most economic value of AI is expected to arise from productivity gains rather than role displacement. We contribute to the automation, augmentation and productivity debates as well as advance our understanding of job redesign in the age of AI.

econ.GN↗

arXiv2014

GRACES: Gemini remote access to CFHT ESPaDOnS Spectrograph through the longest astronomical fiber ever made (Experimental phase completed.)

The Gemini Remote Access to CFHT ESPaDONS Spectrograph has achieved first light of its experimental phase in May 2014. It successfully collected light from the Gemini North telescope and sent it through two 270 m optical fibers to the the ESPaDOnS spectrograph at CFHT to deliver high-resolution spectroscopy across the optical region. The fibers gave an average focal ratio degradation of 14% on sky, and a maximum transmittance of 85% at 800nm. GRACES achieved delivering spectra with a resolution power of R = 40,000 and R = 66,000 between 400 and 1,000 nm. It has a ~8% throughput and is sensitive to target fainter than 21st mag in 1 hour. The average acquisition time of a target is around 10 min. This project is a great example of a productive collaboration between two observatories on Maunakea that was successful due to the reciprocal involvement of the Gemini, CFHT, and NRC Herzberg teams, and all the staff involved closely or indirectly.

astro-ph.IM↗

arXiv2014

Pointwise characteristic factors for Wiener Wintner double recurrence theorem

In this paper, we extend Bourgain's double recurrence result to the Wiener-Wintner averages. Let $(X, \mathcal{F}, μ, T)$ be a standard ergodic system. We will show that for any $f_1, f_2 \in L^\infty(X)$, the double recurrence Wiener-Wintner average \[ \frac{1}{N} \sum_{n=1}^N f_1(T^{an}x)f_2(T^{bn}x) e^{2πi n t} \] converges off a single null set of $X$ independent of $t$ as $N \to \infty$. Furthermore, we will show a uniform Wiener-Wintner double recurrence result: If either $f_1$ or $f_2$ belongs to the orthogonal complement of the Conze-Lesigne factor, then there exists a set of full measure such that the supremum on $t$ of the absolute value of the averages above converges to $0$.

math.DS↗