S8 Ep34: How good are LLMs at doing our jobs?
VoxTalks Economics9 Heinä 2025

S8 Ep34: How good are LLMs at doing our jobs?

 In the second of special series recorded live at the PSE-CEPR Policy Forum 2025, we are asking, how good is AI at doing real-world job task? And how can we measure their capability without resorting to technical benchmarks that may not mean much in the workplace?

Since we all became aware of large language models, LLMs scientists have been attempting to evaluate how good they are at performing expert tasks. The results of those tests can show us whether LLMs can be useful complements to our work, or even replacements for us, as many fear. But setting or grading a test to decide whether an LLM can do a problem-solving job task, rather than solve an abstract problem, isn't easy to do. Maria del Rio-Chanona, a computer scientist at UCL, tells Tim Phillips about her innovative work-in-progress, in which she asks an LLM to set a tricky workplace exam, then tells another LLM to take the test – which a third LLM evaluates.

Tämä jakso on lisätty Podme-palveluun avoimen RSS-syötteen kautta eikä se ole Podmen omaa tuotantoa. Siksi jakso saattaa sisältää mainontaa.

Jaksot(475)

S9 Ep48: What price to save Japan's ghost towns?

S9 Ep48: What price to save Japan's ghost towns?

In the oldest Japanese municipalities, close to half the residents are already over 65. As young people move to cities, the retailers close, then the clinics, then the bus that used to reach the next ...

14 Elo 18min

S9 Ep47: The Next Generation: PSE 2026

S9 Ep47: The Next Generation: PSE 2026

Travel broadens the mind. So the Voxtalks visits a conference, we find the most interesting research from economists just starting out, and hand three of them a microphone. Ad that is today's episode,...

7 Elo 33min

S9 Ep46: Absorbing shocks since 1891

S9 Ep46: Absorbing shocks since 1891

Recorded at the PSE-CEPR Policy Forum 2026.A country wants a stable exchange rate, it wants money to move freely across its borders, and it wants to set its own interest rates. It's a well-known trile...

5 Elo 20min

S9 Ep45: Tariff Confusion

S9 Ep45: Tariff Confusion

If you run a business that exports to the United States, how big is the tariff you have to pay? In 2025 that question was hard to answer. Between February and December, 53 separate announcements intro...

31 Heinä 31min

S9 Ep44: How superpowers shape trade

S9 Ep44: How superpowers shape trade

Recorded at the PSE-CEPR Policy Forum, Paris School of Economics.Tariffs move trade around, but so does economic power. When one economy dominates, other countries fall into step with it. They trade w...

29 Heinä 15min

S9 Ep43: Navigating industrial policy

S9 Ep43: Navigating industrial policy

Recorded at the PSE-CEPR Policy Forum, Paris.In 2019 the IMF called the increasing adoption of industrial policy: "The return of the policy that shall not be named". No one is scared to name it in 202...

24 Heinä 21min

S9 Ep42: The Safety Paradox

S9 Ep42: The Safety Paradox

Recorded at the PSE-CEPR Policy Forum, Paris School of Economics, June 2026.In 1941 the United States banned oil exports to Japan, to punish aggression without fighting a war. Historians now argue the...

22 Heinä 20min

S9 Ep41: Making teenagers read newspapers

S9 Ep41: Making teenagers read newspapers

French teenagers carry a smartphone with access to almost anything, but few of them have been using it to read the news.Julia Cagé (Sciences Po, CEPR) ran an experiment to test the one barrier everyon...

17 Heinä 20min

Suosittua kategoriassa Liike-elämä ja talous

sijotuskasti
mimmit-sijoittaa
psykopodiaa-podcast
rss-rahapodi
hyva-paha-johtaminen
ostan-asuntoja-podcast
rss-porssipuhetta
sijoituspodi
pomojen-suusta
rss-sami-miettinen-neuvottelija
rss-inderes
rss-karon-grilli
rss-ammattipodcast
rss-laakispodi
rss-hilden-kaira
rss-startup-ministerio
yrittaja
salkunrakentaja-podi
rss-vaikuttavan-opettajan-vierella
rss-bisnesta-bebeja