@NCKobis

Professor for Human Understanding of Algorithms and Machines @RCTrustworthy Co-founder of @KickBackGAP & @ICR_Network Basketball- and Anticorruption-nerd

Joined October 2014
Read an interview with the joint first authors Zoe Rahwan @arc_mpib and Nils Köbis @NCKobis about their findings! 👉mpib-berlin.mpg.de/2162756/p…
1
187
Super insightful episode dealing with the very real risks of #syntheticrelationships, esp. among teens. FWIW, we brought together a group of researchers working in that field and propose several policies to address these risks: nature.com/articles/s41562-0… @LaurieSegall @HumaneTech_
It’s Hard Fork Friday! This week on the show, how (and why) Elon Musk made himself the main character of the election. Then, @LaurieSegall joins to talk about the sad case of the 14-year-old boy who took his own life after falling for an AI chatbot.
2
452
🚨 Thrilled to share our latest paper on the risks & protective measures for #SyntheticRelationships, published in @NatureHumBehav🚨 We explore how SRs between humans and AI can pose serious risks to individuals & society. Key concerns include... nature.com/articles/s41562-0… 🧵1/5
1
4
2
11
2,753
🚀Now out in @NatureHumBehav 🚀 How can we reduce dishonest behavior in a tax evasion game? A classic method is through honesty oaths. But… do all honesty oaths work the same? To answer this question, we adopted a mega-study approach, testing 21 different honesty oaths on more than 20.000 people. The results are quite surprising: both experts and laypeople failed to predict the most effective interventions. Some oaths worked, others didn’t. The most effective oath almost halved tax evasion. So, which oath was it? It was one that highlighted specific behavior: “I hereby declare that I will provide honest information when reporting my final income from the sorting task.” You can read the full paper here: nature.com/articles/s41562-0… This project was wonderfully led by Janis Zickfield: @Goatfield2 w/ @kascigala, @CElbaek, John Michael, Mathilde Tonnesen, @gabriellevy1, Shahar Ayal, @IsabelThielmann, @LailaNockur, @E_yal_Pe_er, Rachel Barkan, Simen Bo, @bahniks, Daniele Nosenzo, Ralph Hertwig, Nina Mazar, Alexa Weiss, Ann-Kathrin Koessler, @MontalRoni51916, @shafenbraedl, Yngwie Asbjorn Nielsen, @PatKanngiesser, Simon Schindler, Philipp Gerlach, @NCKobis, Nicolas Jacquement, @mVranka, @danariely, @jareefbnmartuza, @feldman_yuval, @mbialek82, @jan_k_woike, Zoe Rahwan, @alicia_seidl, Eileen Chou, @agnekaj, @simeonschudy, @UGlogowsky, Anna Czarna, Stefan Pfattheicher, Panagiotis Mitkidis.
3
20
3
64
11,134
Lie-detection AI could provoke people into making careless accusations, researchers warn. Read more in @iScience_CP: cell.com/iscience/fulltext/S… @NCKobis
1
7
3,218
L'IA : outil révolutionnaire contre la corruption ou à double tranchant? Ne manquez pas notre épisode avec @NCKobis, explorant les opportunités et les défis. 🎧 Episode en VO et en VF à retrouver sur toutes les plateformes #IA #Éthique #corruption buff.ly/3VBuFWh
2
130
One week left to apply for our Post-Doc at the @RCTrustworthy @uniallianz Wilhelm Hofmann (@ruhrunibochum) , Margarita Leib (@TilburgU) & I are looking for a PostDoc to work on people’s perceptions of and interactions with #AI in daily lives academics.com/jobs/post-doct… #academicjobs
2
4
636
Excited about our #EADMSummerSchool24 on Large Language Models (LLMs) in Behavioural Science! ⏰ Application deadline is less than a month away (March 17th) Besides the magic of Berlin in late summer, you’d get:
1
13
25
3,939
Nils Köbis retweeted
Join us at this year's @EADM_1993 summer school hosted by @arc_mpib and @Max_Planck_CHM at @mpib_berlin, Sep 9-13, with keynotes from @MelMitchell1, @iyadrahwan, @gael_lemens, and @longoni_chiara and talks from @weidingerlaura, @jason_w_burton, @Zoe_A_Purcell, @NCKobis!
1
12
1
25
5,214
An absolute honor to be invited to present alongside @DAcemogluMIT, @FryRsquared, @PetterBB, @alisonmdarcy about the risks of AI corrupting human ethical behavior review: nature.com/articles/s41562-0… at the 20th European Trend Day (@GDInstitute). Hopefully see you in Zurich!
1
18
650
📍We're thrilled to open a call for papers to the 8th #Interdisciplinary #Corruption #Research Network ICRN Forum 2024! 🏠Where? the Institute of Social Sciences, the University of Lisbon, ⏲️When? June 27 - 29, 2024. Apply by: ⏳ February 9, 2024 👉  lnkd.in/dsEmQazn
1
9
3
13
2,649
La corruption : un mal nécessaire ? | 42, la réponse à presque tout | ARTE youtu.be/1rxatiYun20?si=IbSw… via @YouTube German version : arte.tv/de/videos/109818-008… Very exciting to participate with @NCKobis, @HussOksana, and Ivo Engels!
1
1
2
586
🚨Call for papers: Meta-analysis on the value of Algorithmic advice We are conducting a meta-analysis assessing how people value advice and input given by algorithms. We refer to advice as a recommendation favoring or discouraging particular option(s).
1
5
7
1,172
Our new piece on the Behavioral Ethics of #AI is out now in @PsychToday It covers whether #Transparency about the involvement of algorithms is enough to reduce AI's corrupting influence on human ethical behavior. psychologytoday.com/intl/blo… #AIEthics
1
3
2
16
8,220
New paper in J of Behavioral and Experimental Economics. How different incentive schemes affect corruption. With Soraperra, @NCKobis, Shalvi, @sb_vogt, and Offerman. doi.org/10.1016/j.socec.2023…
4
6
646
2) Inequality aversion: People don't differentiate between humans and machines in behindness aversion, indicating partner-independent envy. For aheadness aversion, humans and humans behind machines trigger more aversion compared to machines, suggesting no sympathy for machines.
1
1
2
181
3) Positive and negative reciprocity is more pronounced when dealing with humans and humans behind machines compared to machines; there is a reluctance to reciprocate with machines. 4) Lack of information about machine payoffs leads to motivated reasoning.
1
1
176