AI-Generated Music Detection and its Challenges

Darius Afchar; Gabriel Meseguer-Brocal; Romain Hennequin

REVIEW 4 major objections 5 minor 35 references

AI-Generated Music Detection and its Challenges

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A simple convolutional classifier trained on real tracks versus their autoencoder reconstructions detects AI-made music at 99.8% accuracy, but the detector collapses under pitch shifts, added noise, and re-encoding, and only generalizes…

desk verdict An honest, useful first study whose real contributions are the negative results, not the 99.8% headline; the MusicGen transfer test is confounded by the shared Encodec decoder. read the letter →

arxiv 2501.10111 v1 pith:GNU25MSE submitted 2025-01-17 cs.SD eess.AS

Darius Afchar , Gabriel Meseguer-Brocal , Romain Hennequin This is my paper · ORCID

classification cs.SDeess.AS

keywords AI-generatedmusicdetectiondeepfakeaudioforensicsneuralcodecsautoencoderfingerprintingtext-to-musicgenerationrobustnesstomanipulationdetectorgeneralizationconvolutionalnetwork

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

AI-generated music can be detected with surprising ease: a basic convolutional network that compares a real track with its own reconstruction through a generator's autoencoder reaches 99.8% accuracy on amplitude spectrograms. The paper's second, cautionary finding is that this ability is narrow. The same detector scores near zero when audio is pitch-shifted, mixed with white noise, or re-encoded, and a model trained on one autoencoder family fails almost completely on another. The authors present this as the first published AI-music detector while arguing that a high test score does not make a deployable tool: performance must be measured against manipulations, unseen generators, calibration, and the risk of a model that simply defaults to 'real' whenever it does not find a familiar artefact.

What carries the argument

The carrying object is the decoder fingerprint: the small reconstruction artefacts left by the autoencoder stage of waveform-based music generators, such as the checkerboard patterns associated with transposed convolutions. The protocol controls for confounds by taking a real track and passing it through the generator's autoencoder at the same bitrate, so the only difference between the two examples is the decoder's artefact. A six-layer convolutional network with 16 to 512 filters, trained on random 0.8-second amplitude-spectrogram snippets with a one-in-two real-versus-synthetic sampling ratio, is enough to exploit this difference. The same mechanism explains the failures: each autoencoder family leaves a different fingerprint, and the model's bias toward simple solutions turns the real class into a default category rather than a learned manifold of authentic music.

What would settle it

Train the released detector on real audio versus reconstructions from only GriffinMel and DAC (neither is Encodec), then evaluate it on full MusicGen generations; if accuracy falls to near chance, the Section II.D 99.9% score came from codec familiarity rather than from a general generation artefact, and the central claim is autoencoder fingerprinting, not AI-music detection.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that AI-music detection can be reduced to autoencoder fingerprinting. By autoencoding 25,000 real tracks from the Free Music Archive (FMA) with nine configurations of four decoders (the neural codecs Encodec and DAC, the GriffinMel mel-spectrogram pipeline, and the Musika decoder) and training a six-layer CNN on 0.8-second amplitude-spectrogram snippets, the authors obtain 99.8% test accuracy and 99.9% accuracy on 2,500 snippets from the full text-to-music generator MusicGen. The catch documented in the same paper is that the fingerprint is not a universal marker of machine generation: detection transfers within the Encodec family (24 kbps training covers 6 and 3 kbps) but is near zero across families, such as GriffinMel to DAC, and common manipulations like a two-semitone pitch shift and 64 kbps re-encoding drop scores to near zero because the model predicts 'real' when it encounters no known artefact. The authors' claim is therefore twofold: forged music is easy to detect in a matched setting, and that ease is precisely why the detector should not be trusted outside it.

Load-bearing premise

The load-bearing premise is that running a real track through a generator's autoencoder leaves exactly the same artefacts as that generator's actual output, so a classifier trained on these reconstructions is a valid detector of real AI-generated music; the paper checks this against only one full generator, MusicGen, which shares Encodec with the training data.

Editorial extensions

If this is right

A detector trained on one bitrate of a neural codec family transfers to other bitrates of the same family, so not every codec configuration needs to be seen during training.
Common user-level manipulations, including pitch shifting by two semitones, adding white noise, or re-encoding to mp3, AAC, or Opus at 64 kbps, can reduce accuracy to near zero, making a naively deployed detector unsafe.
Fine-tuning on manipulations and on new decoders can restore high accuracy, but since unseen manipulations and generators will keep appearing, detection needs a continual patching process rather than a one-time model.
The model's tendency to default to the real class means its output probabilities should not be interpreted as a percentage of AI content, which matters for calibration and for mixed real-and-synthetic stems.
The high score on MusicGen supports autoencoder fingerprinting rather than proof of generalization, because MusicGen shares the Encodec family seen during training.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Our inference: the matched-reconstruction protocol could become a standard evaluation scheme in which every reported accuracy is paired with a cross-family score, so a headline number like 99.8% is never quoted without its generalization envelope.
Our inference: because inter-family scores are near zero, combining many codec families in one training set may or may not yield a detector that learns a common neural-codec artefact; the paper does not test this, and it is the natural next experiment.
Our inference: the near-zero performance under pitch shift and re-encoding suggests that resistance to manipulation and to new generators will be bought with data augmentation and continual fine-tuning, and a practical detector should be evaluated under a partial-knowledge protocol that simulates an attacker who knows the detector.
Our inference: the paper's framing implies that content provenance measures such as watermarking and platform-level regulation may prove more scalable than chasing every new generator with a detector, even though the paper notes that watermarking is itself brittle.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

This is the first paper I know of that tackles instrumental-music deepfake detection in a controlled way. The setup is sensible: take real FMA tracks, autoencode them with nine decoder configurations, and train a simple CNN to tell original from reconstruction. The 99.8% number checks out internally, and the cross-decoder transfer matrix (Table III) is the most informative thing in the paper. It shows intra-family generalization (Encodec 24kbps to 3kbps works) and inter-family collapse (everything goes to zero). The authors deserve credit for publishing that—most papers would bury it.

The paper is also refreshingly self-critical. They explicitly say the model is probably detecting decoder artefacts, that the real class acts as a default, and that fine-tuning will always be chasing an unseen generator. That framing is more honest than the abstract, which claims "the first AI-music detector" and implies general detection.

Now the soft spots. The stress-test concern about MusicGen is valid: MusicGen uses Encodec, which dominates the training set, so the 99.9% transfer is the expected result if the model has learned Encodec-specific fingerprints. It does not validate the proxy assumption for non-Encodec generators. The authors list VampNet, MusicLM, and Jukebox as motivating models but test none of them. One non-Encodec full generator would have settled the question. Related: Table II is rendered without any numbers in the arXiv PDF, so the robustness results cannot actually be read. That is an editorial fix, but a referee should insist on it. There are no confidence intervals either, though with 2500 snippets the CIs would be tight except where accuracy is near zero or one.

The bigger conceptual issue is that the central claim of AI-music detection is only supported within the Encodec family. The paper's own evidence shows that a detector trained on Encodec does not detect DAC, GriffinMel, or Musika outputs. So the title and abstract overstate. That said, the authors already say most of this in Sections III.A and III.B—the problem is mostly presentation.

Who is this for? Anyone building or evaluating music deepfake detectors, and anyone studying shortcut learning in audio classification. It is a proof of concept with strong negative results, not a deployable tool. I would send it to a serious referee, and I would ask the authors to add one cross-family generator test, fix Table II, and temper the abstract. I would cite it for the transfer matrix and the confound-controlled benchmark.

My verdict: conditional accept, with the Table II repair and a non-Encodec generalization test as conditions. The core honesty of the paper makes it a useful contribution even if the headline claim does not fully survive.

Referee Report

4 major / 5 minor

Summary. The paper proposes detecting AI-generated music by training simple CNNs to distinguish real audio from autoencoder reconstructions of the same tracks, using nine reconstructions from Encodec, DAC, GriffinMel, and Musika applied to the FMA dataset. The amplitude-spectrogram model reaches 99.8% held-out accuracy, and the authors report 99.9% accuracy on 50 MusicGen tracks. The second half documents that accuracy collapses under common audio manipulations and that models trained on one decoder family do not generalize to other families, which the authors present as caveats for deployed detectors. The paper is framed both as a first AI-music detector and as a position paper on the challenges of such detectors.

Significance. If the core result stands, it is a valuable proof-of-concept that known decoder families leave detectable artefacts, and the paper's self-reported negative results—especially the cross-family collapse in Table III and the "default to real" behavior—are useful cautionary evidence for the emerging field of AI-music detection. The authors also make code and trained weights available, which supports reproducibility. However, the central generalization claim from autoencoder reconstructions to real AI-generated music is only validated within the Encodec family, since MusicGen uses Encodec as its decoder; the advertised "AI-music detector" is therefore broader than the evidence supports. The missing numerical content of Table II and the absence of confidence intervals further weaken the robustness discussion.

major comments (4)

[II.D and II.B] The 99.9% MusicGen transfer result does not establish detection of AI-generated music in general, because MusicGen uses Encodec as its audio decoder, as the authors themselves note in Section II.B. Since Encodec reconstructions are in the training set, the MusicGen test is an in-family generalization test, not a test of detection across generator families. Table III makes this concern concrete: an Encodec-trained model scores 0% on DAC and GriffinMel. The claim in the abstract and Section II.D should be narrowed to "detection of autoencoder artefacts" and "transfer to an Encodec-based generator," or the paper should add a full-generator evaluation using a non-Encodec decoder (e.g., VampNet/DAC, MusicLM/SoundStream, or Jukebox/VQ-VAE).
[Table II] Table II, which supports the robustness claims in Section III.A, contains no readable numeric values in the manuscript: only the caption and a table frame appear, with the actual accuracy numbers missing. The claims about pitch-shift, noise, and codec-reencoding performance collapsing to near zero cannot be checked without those values. Please provide the complete table, including per-class breakdowns and confidence intervals or per-run variability.
[Section III.C] The statement that "fine-tuning on these manipulations could reliably restore high accuracy scores" and "the same is true about fine-tuning to a new decoder" is not supported by any experiment shown in the paper. Since this is offered as a mitigation direction, either include the corresponding experiments with numbers, or explicitly label it as an unreported preliminary observation.
[Section II.C and Table I] No confidence intervals or statistical significance tests are reported for the accuracies in Table I. With a test split of 20% of 25,000 tracks, the differences among 99.8%, 99.7%, and 99.6% may be within noise; reporting standard errors or confidence intervals would clarify whether the amplitude representation is genuinely better than phase or polar representations.

minor comments (5)

[Abstract] "a AI-music detector" should be "an AI-music detector."
[Section II.B] The name "Mo ˆusai" appears with corrupted formatting; it should be rendered as "Mousai" or "Moûsai" consistently.
[Section II.C] "sampled with a 1 2 probability" should read "sampled with probability 1/2."
[Section II.D] The MusicGen test uses only 50 tracks (25 minutes of audio). While the snippet count is 2500, these snippets are not independent because they come from only 50 generated tracks; a track-level analysis or a larger corpus would strengthen the claim.
[Table III] The table is difficult to parse in text form; a heatmap or a matrix with explicit row/column labels would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the reported accuracies are empirical measurements and the AE-fingerprint proxy is disclosed, with cross-family generalization failures explicitly reported.

full rationale

The paper's 99.8% accuracy is an empirical result on a held-out test split of FMA audio versus nine autoencoder reconstructions, not a derived quantity or a fitted parameter relabeled as a prediction. The central framework is stated transparently: the model is trained to distinguish real audio from its autoencoded reconstruction, and the paper explicitly says it aims to detect 'the fingerprint of the AE' rather than the generator itself (Section II.D). The MusicGen transfer test shares the Encodec decoder used in training, so it does not independently validate detection of unseen generator families; however, the paper itself reports this limitation through Table III, showing near-zero inter-family generalization, and Section III.A explains that the model defaults to the real class when familiar artefacts are absent. This is a validity or confound concern, not a definitional circularity: the training labels are real versus reconstructed audio, and no target quantity defined in terms of the claimed prediction is used to construct the inputs. There are no load-bearing self-citations, no imported uniqueness theorems, no ansatz smuggled in via citation, and no renaming of a known result. The missing Table II values and the absence of confidence intervals affect reproducibility but do not make the derivation circular. Overall, the paper's honest caveats about robustness and generalization keep the circularity burden low.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The paper's headline accuracy is not a derived quantity, but it is conditional on hand-chosen experimental settings: snippet length, frequency cutoff, and the specific decoder set. The key domain assumptions are the confounding-control framework and the AE-fingerprint proxy, both explicitly acknowledged in Sections II.A and II.D.

free parameters (3)

Snippet length = 0.8 s (128 STFT frames)
Chosen arbitrarily; determines the time context seen by the classifier and affects the reported accuracy.
STFT frequency cutoff = 16 kHz
Hand-picked to avoid relying on mp3 conversion artefacts; changes what the model can exploit.
Decoder set (nine reconstructions) = Encodec 3/6/24 kbps, DAC 2/7/14 kbps, GriffinMel 256/512, Musika
Dataset composition manually selected; the cross-family results in Table III depend on this selection.

assumptions (3)

domain assumption Autoencoding a real track with an AE and comparing it to the original isolates generation artefacts while controlling for musical content and bitrate.
Section II.A. The entire confounding-control framework rests on this premise; if bitrate or semantic confounds remain, high accuracy could come from non-artefact cues.
domain assumption Detecting the AE fingerprint is sufficient to detect full AI-music generators, because the AE performs waveform synthesis in these models.
Sections II.A and II.D. Only tested on MusicGen, which is Encodec-based; not established for other full generators such as Jukebox or MusicLM.
domain assumption The real class is treated by the model as a default when no known artefact is present.
Sections III.A and III.B. Used to explain zero-accuracy scores under manipulation and cross-family transfer.

how reviews work

0 comments

Cite this review

Pith. "Pith review of AI-Generated Music Detection and its Challenges." pith.science (2026). https://pith.science/paper/GNU25MSE

@misc{pith2026250110111,
  author       = {Pith},
  title        = {Pith review of: AI-Generated Music Detection and its Challenges},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GNU25MSE}},
  note         = {Machine review of arXiv:2501.10111}
}

read the original abstract

In the face of a new era of generative models, the detection of artificially generated content has become a matter of utmost importance. In particular, the ability to create credible minute-long synthetic music in a few seconds on user-friendly platforms poses a real threat of fraud on streaming services and unfair competition to human artists. This paper demonstrates the possibility (and surprising ease) of training classifiers on datasets comprising real audio and artificial reconstructions, achieving a convincing accuracy of 99.8%. To our knowledge, this marks the first publication of a AI-music detector, a tool that will help in the regulation of synthetic media. Nevertheless, informed by decades of literature on forgery detection in other fields, we stress that getting a good test score is not the end of the story. We expose and discuss several facets that could be problematic with such a deployed detector: robustness to audio manipulation, generalisation to unseen models. This second part acts as a position for future research steps in the field and a caveat to a flourishing market of artificial content checkers.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 28 canonical work pages

[1]

Watching the generative ai hype bubble deflate,

David Gray Widder and Mar Hicks, “Watching the generative ai hype bubble deflate,” arXiv:2408.08778, 2024

arXiv 2024
[2]

Data science as political action: Grounding data science in a politics of justice,

Ben Green, “Data science as political action: Grounding data science in a politics of justice,” Journal of Social Computing , vol. 2, no. 3, 2021

work page 2021
[3]

Kate Crawford, The atlas of AI: Power, politics, and the planetary costs of artificial intelligence , Yale University Press, 2021

work page 2021
[4]

Ai art is theft: Labour, extraction, and exploitation: Or, on the dangers of stochastic pollocks,

Trystan S Goetze, “Ai art is theft: Labour, extraction, and exploitation: Or, on the dangers of stochastic pollocks,” in ACM FAccT, 2024

work page 2024
[5]

From melting pots to misrepresentations: Exploring harms in generative ai,

Sanjana Gautam, Pranav Narayanan Venkit, and Sourojit Ghosh, “From melting pots to misrepresentations: Exploring harms in generative ai,” in GenAICHI, 2024

work page 2024
[6]

Ai art and its impact on artists,

Harry H. Jiang, Lauren Brown, Jessica Cheng, Mehtab Khan, Abhishek Gupta, Deja Workman, Alex Hanna, Johnathan Flowers, and Timnit Gebru, “Ai art and its impact on artists,” in AIES. 2023, ACM

work page 2023
[7]

Moˆusai: Efficient text-to-music diffusion models,

Flavio Schneider, Ojasv Kamal, Zhijing Jin, and Bernhard Sch ¨olkopf, “Moˆusai: Efficient text-to-music diffusion models,” ACL, 2024

work page 2024
[8]

Musiclm: Generating music from text,

Andrea Agostinelli, Timo I Denk, Zal ´an Borsos, Jesse Engel, Mauro Verzetti, Antoine Caillon, Qingqing Huang, Aren Jansen, Adam Roberts, Marco Tagliasacchi, et al., “Musiclm: Generating music from text,” arXiv:2301.11325, 2023

arXiv 2023

Show all 35 references

[9]

Vampnet: Music generation via masked acoustic token mod- eling,

Hugo Flores Garcia, Prem Seetharaman, Rithesh Kumar, and Bryan Pardo, “Vampnet: Music generation via masked acoustic token mod- eling,” ISMIR, 2023

2023
[10]

Simple and controllable music generation,

Jade Copet, Felix Kreuk, Itai Gat, Tal Remez, David Kant, Gabriel Synnaeve, Yossi Adi, and Alexandre D´efossez, “Simple and controllable music generation,” NeurIPS, vol. 36, 2024

2024
[11]

Asvspoof: the automatic speaker verifica- tion spoofing and countermeasures challenge,

Zhizheng Wu, Junichi Yamagishi, Tomi Kinnunen, Cemal Hanilc ¸i, Mohammed Sahidullah, Aleksandr Sizov, Nicholas Evans, Massimiliano Todisco, and Hector Delgado, “Asvspoof: the automatic speaker verifica- tion spoofing and countermeasures challenge,” IEEE Journal of Selected Top...

2017
[12]

Singfake: Singing voice deepfake detection,

Yongyi Zang, You Zhang, Mojtaba Heydari, and Zhiyao Duan, “Singfake: Singing voice deepfake detection,” in ICASSP. IEEE, 2024

2024
[13]

Faceforensics++: Learning to detect manipulated facial images,

Andreas Rossler, Davide Cozzolino, Luisa Verdoliva, Christian Riess, Justus Thies, and Matthias Nießner, “Faceforensics++: Learning to detect manipulated facial images,” in ICCV. 2019, IEEE

2019
[14]

The creation and detection of deepfakes: A survey,

Yisroel Mirsky and Wenke Lee, “The creation and detection of deepfakes: A survey,” ACM computing surveys (CSUR) , vol. 54, no. 1, 2021

2021
[15]

Detecting multimedia generated by large ai models: A survey,

Li Lin, Neeraj Gupta, Yue Zhang, Hainan Ren, Chun-Hao Liu, Feng Ding, Xin Wang, Xin Li, Luisa Verdoliva, and Shu Hu, “Detecting multimedia generated by large ai models: A survey,” arXiv:2402.00045, 2024

2024 arXiv
[16]

Melgan: Generative adversarial networks for conditional waveform synthesis,

Kundan Kumar, Rithesh Kumar, Thibault De Boissiere, Lucas Gestin, Wei Zhen Teoh, Jose Sotelo, Alexandre De Brebisson, Yoshua Bengio, and Aaron C Courville, “Melgan: Generative adversarial networks for conditional waveform synthesis,” NeurIPS, vol. 32, 2019

2019
[17]

Hifi-gan: Generative adversarial networks for efficient and high fidelity speech synthesis,

Jungil Kong, Jaehyeon Kim, and Jaekyoung Bae, “Hifi-gan: Generative adversarial networks for efficient and high fidelity speech synthesis,” NeurIPS, vol. 33, 2020

2020
[18]

Jukebox: A generative model for music,

Prafulla Dhariwal, Heewoo Jun, Christine Payne, Jong Wook Kim, Alec Radford, and Ilya Sutskever, “Jukebox: A generative model for music,” arXiv:2005.00341, 2020

2005 arXiv
[19]

Musika! fast infinite waveform music generation,

Marco Pasini and Jan Schl ¨uter, “Musika! fast infinite waveform music generation,” in ISMIR, 2022

2022
[20]

Soundstream: An end-to-end neural audio codec,

Neil Zeghidour, Alejandro Luebs, Ahmed Omran, Jan Skoglund, and Marco Tagliasacchi, “Soundstream: An end-to-end neural audio codec,” IEEE/ACM Transactions on Audio, Speech, and Language Processing , vol. 30, 2021

2021
[21]

High fidelity neural audio compression,

Alexandre D ´efossez, Jade Copet, Gabriel Synnaeve, and Yossi Adi, “High fidelity neural audio compression,” arXiv:2210.13438, 2022

2022 arXiv
[22]

High-fidelity audio compression with improved rvqgan,

Rithesh Kumar, Prem Seetharaman, Alejandro Luebs, Ishaan Kumar, and Kundan Kumar, “High-fidelity audio compression with improved rvqgan,” NeurIPS, vol. 36, 2024

2024
[23]

Generating diverse high-fidelity images with vq-vae-2,

Ali Razavi, Aaron Van den Oord, and Oriol Vinyals, “Generating diverse high-fidelity images with vq-vae-2,” NeurIPS, vol. 32, 2019

2019
[24]

Jonas Peters, Dominik Janzing, and Bernhard Sch ¨olkopf, Elements of causal inference: foundations and learning algorithms , The MIT Press, 2017

2017
[25]

FMA: A dataset for music analysis,

Micha ¨el Defferrard, Kirell Benzi, Pierre Vandergheynst, and Xavier Bresson, “FMA: A dataset for music analysis,” in ISMIR, 2017

2017
[26]

As good as a coin toss human detection of ai-generated images, videos, audio, and audiovisual stimuli,

Di Cooke, Abigail Edwards, Sophia Barkoff, and Kathryn Kelly, “As good as a coin toss human detection of ai-generated images, videos, audio, and audiovisual stimuli,” arXiv:2403.16760, 2024

2024 arXiv
[27]

Why ai writing detectors don’t work,

Benj Edwards, “Why ai writing detectors don’t work,” Ars Technica – https://arstechnica.com, 2023

2023
[28]

Black-box access is insufficient for rigorous ai audits,

Stephen Casper, Carson Ezell, Charlotte Siegmann, Noam Kolt, Tay- lor Lynn Curtis, Benjamin Bucknall, Andreas Haupt, Kevin Wei, J ´er´emy Scheurer, Marius Hobbhahn, et al., “Black-box access is insufficient for rigorous ai audits,” in ACM FAccT, 2024

2024
[29]

Cnn-generated images are surprisingly easy to spot... for now,

Sheng-Yu Wang, Oliver Wang, Richard Zhang, Andrew Owens, and Alexei A Efros, “Cnn-generated images are surprisingly easy to spot... for now,” in ICCV, 2020

2020
[30]

Celeb- df: A large-scale challenging dataset for deepfake forensics,

Yuezun Li, Xin Yang, Pu Sun, Honggang Qi, and Siwei Lyu, “Celeb- df: A large-scale challenging dataset for deepfake forensics,” in ICCV, 2020

2020
[31]

The in- formation bottleneck method,

Naftali Tishby, Fernando C Pereira, and William Bialek, “The in- formation bottleneck method,” 37th annual Allerton Conference on Communication, Control, and Computing , 2000

2000
[32]

Which shortcut cues will dnns choose? a study from the parameter-space perspective,

Luca Scimeca, Seong Joon Oh, Sanghyuk Chun, Michael Poli, and Sangdoo Yun, “Which shortcut cues will dnns choose? a study from the parameter-space perspective,” in ICLR, 2021

2021
[33]

Tackling deepfakes in european policy,

Djurre Das, Pieter van Boheemen, Nierling Linda, Jutta Jahnel, Murat Karaboga, Martin Fatun, and Mari ¨ette Huijstee, “Tackling deepfakes in european policy,” Tech. Rep., European Parliament, 07 2021

2021
[34]

Combatting deepfakes: Policies to ad- dress national security threats and rights violations,

Andrea Miotti and Akash Wasil, “Combatting deepfakes: Policies to ad- dress national security threats and rights violations,” arXiv:2402.09581, 2024

2024 arXiv
[35]

The brittleness of ai-generated image watermark- ing techniques: Examining their robustness against visual paraphrasing attacks,

Niyar R Barman, Krish Sharma, Ashhar Aziz, Shashwat Bajpai, Shwe- tangshu Biswas, Vasu Sharma, Vinija Jain, Aman Chadha, Amit Sheth, and Amitava Das, “The brittleness of ai-generated image watermark- ing techniques: Examining their robustness against visual paraphrasing attack...

2024 arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

Watching the generative ai hype bubble deflate,

David Gray Widder and Mar Hicks, “Watching the generative ai hype bubble deflate,” arXiv:2408.08778, 2024

arXiv 2024

[2] [2]

Data science as political action: Grounding data science in a politics of justice,

Ben Green, “Data science as political action: Grounding data science in a politics of justice,” Journal of Social Computing , vol. 2, no. 3, 2021

work page 2021

[3] [3]

Kate Crawford, The atlas of AI: Power, politics, and the planetary costs of artificial intelligence , Yale University Press, 2021

work page 2021

[4] [4]

Ai art is theft: Labour, extraction, and exploitation: Or, on the dangers of stochastic pollocks,

Trystan S Goetze, “Ai art is theft: Labour, extraction, and exploitation: Or, on the dangers of stochastic pollocks,” in ACM FAccT, 2024

work page 2024

[5] [5]

From melting pots to misrepresentations: Exploring harms in generative ai,

Sanjana Gautam, Pranav Narayanan Venkit, and Sourojit Ghosh, “From melting pots to misrepresentations: Exploring harms in generative ai,” in GenAICHI, 2024

work page 2024

[6] [6]

Ai art and its impact on artists,

Harry H. Jiang, Lauren Brown, Jessica Cheng, Mehtab Khan, Abhishek Gupta, Deja Workman, Alex Hanna, Johnathan Flowers, and Timnit Gebru, “Ai art and its impact on artists,” in AIES. 2023, ACM

work page 2023

[7] [7]

Moˆusai: Efficient text-to-music diffusion models,

Flavio Schneider, Ojasv Kamal, Zhijing Jin, and Bernhard Sch ¨olkopf, “Moˆusai: Efficient text-to-music diffusion models,” ACL, 2024

work page 2024

[8] [8]

Musiclm: Generating music from text,

Andrea Agostinelli, Timo I Denk, Zal ´an Borsos, Jesse Engel, Mauro Verzetti, Antoine Caillon, Qingqing Huang, Aren Jansen, Adam Roberts, Marco Tagliasacchi, et al., “Musiclm: Generating music from text,” arXiv:2301.11325, 2023

arXiv 2023