The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: ArrowInvalid
Message: JSON parse error: The document is empty.
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 324, in _generate_tables
df = pandas_read_json(f)
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 38, in pandas_read_json
return pd.read_json(path_or_buf, **kwargs)
~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/pandas/io/json/_json.py", line 791, in read_json
json_reader = JsonReader(
path_or_buf,
...<16 lines>...
engine=engine,
)
File "/usr/local/lib/python3.14/site-packages/pandas/io/json/_json.py", line 905, in __init__
self.data = self._preprocess_data(data)
~~~~~~~~~~~~~~~~~~~~~^^^^^^
File "/usr/local/lib/python3.14/site-packages/pandas/io/json/_json.py", line 917, in _preprocess_data
data = data.read()
File "/usr/local/lib/python3.14/site-packages/datasets/utils/file_utils.py", line 844, in read_with_retries
out = read(*args, **kwargs)
File "<frozen codecs>", line 325, in decode
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xbf in position 23: invalid start byte
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2840, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 327, in _generate_tables
raise e
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 290, in _generate_tables
pa_table = paj.read_json(
io.BytesIO(batch), read_options=paj.ReadOptions(block_size=block_size)
)
File "pyarrow/_json.pyx", line 342, in pyarrow._json.read_json
File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
raise convert_status(status)
pyarrow.lib.ArrowInvalid: JSON parse error: The document is empty.Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
claim-registries
Claim-maintenance registries (measurement, not certification), mirrored byte-for-byte from
https://councilof.ai/claims/, each with its OpenTimestamps receipt beside it.
No claim of falsity is made about any entry. A registry records what a named party's own
public record says, what our harness measured, the window and denominator of each reading, and —
for every claim — an explicit list of what the reading does not establish. Where a search
could not reach an organisation's own record, that is recorded as SEARCH_INCONCLUSIVE and never
as an absence. Nothing here is sent to any party named in it. No financial advice, no valuation,
no solvency or adequacy opinion.
Contents
ondo-chainlink/ — eight claims captured from chain.link and ondo.finance on 2026-09-22.
| File | What it is |
|---|---|
claimreg-ondo-chainlink-2026-09-22.json |
rev1: the eight claims as captured, all CLAIM_CAPTURED, each with a measurement plan. Superseded, never edited. |
claimreg-ondo-chainlink-2026-09-22-rev2.json |
rev2: the same eight after the measurements ran. 5 CLAIM_MEASURED, 2 UNMEASURED, 1 CLAIM_CAPTURED. |
claimreg-ondo-chainlink-2026-09-22-rev2.signed.json |
the Ed25519 signature over rev2's bytes by did:web:csoai.org#board-attestation-1. A signature cannot live inside the bytes it covers, so it is a sidecar that pins the file by sha256. |
series/CL-1.jsonl |
the published dated series for a cumulative counter that cannot be recomputed. One reading is not a measurement; the claim stays UNMEASURED until the series holds two dates. |
findings-ondo-chainlink-2026-09-22.md |
the readable note: what was measured, what was found, what was explicitly not established, and how to re-run it. |
*.ots |
OpenTimestamps receipts. These are calendar-pending: a calendar's promise of future Bitcoin inclusion, which is not a Bitcoin attestation and is not described as one until upgraded and verified. |
rev2 supersedes rev1 by reference: rev1's bytes are unchanged, still served at their own URL and still covered by their own receipt.
Verifying a registry
The eight claim records in rev2 are committed to an RFC 9162 §2.1.1 Merkle root (0x00 leaf / 0x01 node, largest-power-of-two split, no odd-leaf duplication), with an inclusion proof for each record in the file — so any single claim can be proved to have been in this registry without republishing the rest.
# root and inclusion proofs
import json, hashlib
d = json.load(open("claimreg-ondo-chainlink-2026-09-22-rev2.json"))
leaf = lambda b: hashlib.sha256(b"\x00" + b).digest()
node = lambda l, r: hashlib.sha256(b"\x01" + l + r).digest()
canon = lambda o: json.dumps(o, sort_keys=True, separators=(",", ":"), ensure_ascii=False).encode()
recs = {c["id"]: dict(c, subject=s) for s, v in d["subjects"].items() for c in v["claims"]}
entries = [canon(recs[l["id"]]) for l in d["merkle"]["leaves"]]
def mth(e):
if len(e) == 1: return leaf(e[0])
k = 1 << (len(e) - 1).bit_length() - 1
return node(mth(e[:k]), mth(e[k:]))
assert mth(entries).hex() == d["merkle"]["root"]
The signature verifies against #board-attestation-1 in
https://csoai.org/.well-known/did.json; the sidecar carries the exact canonicalisation rule.
Re-running the measurements
The harness is scripts/claims/ in https://github.com/CSOAI-ORG/councilof-ai. Python standard
library only, and keyless by construction rather than by promise: the fetch helper raises if a
harness tries to send an Authorization header, an API key or a cookie, and a source that answers
HTTP 402 is recorded as paid and dropped rather than worked around.
python3 scripts/claims/test_claim_harness.py # 21 tests
python3 scripts/claims/test_claim_harness.py --controls # the planted input each harness must reject
python3 scripts/claims/run_all.py /tmp/run
python3 scripts/claims/build_rev2.py /tmp/run /tmp/rev.json
Every measurable claim type ships with a control that proves it can fail — a doubled heartbeat
gap, an implausible on-chain timestamp, 99% of a slice that is 1% of the market, a counter that
goes down, a term that exists only inside a <script> tag. A harness that has never failed has
not been tested.
Licence: CC-BY-4.0. We measure; we never certify.
Objections, contact and corrections
To object to a row, ask for a re-check, request a correction or ask for a record to be withdrawn, email nicholas@csoai.org or use the appeals and dispute route at https://councilof.ai/dispute/. Corrections are listed in the corrections ledger at https://councilof.ai/api/corrections, with what changed and when. CSOAI Ltd (company no. 16939677, England and Wales) is the accountable publisher.
How to cite
CSOAI Ltd (Council of AI). claim-registries. 2026. Hugging Face dataset csoai/claim-registries. https://huggingface.co/datasets/csoai/claim-registries
@misc{csoai_claim_registries,
title = {claim-registries},
author = {{CSOAI Ltd}},
year = {2026},
howpublished = {Hugging Face dataset, https://huggingface.co/datasets/csoai/claim-registries},
note = {Corrections: https://councilof.ai/api/corrections}
}
Licence: CC-BY-4.0. Attribute Council of AI, CSOAI Ltd (16939677), https://councilof.ai.
Corrections and verification
- Corrections ledger (signed): https://councilof.ai/api/corrections. Corrections to CSOAI's published records are logged there with what changed and when.
- Verify a signed record yourself, free and without an account: https://councilof.ai/gspc-verify/ (step by step: https://councilof.ai/signed/HOW-TO-VERIFY.md).
- Conformance kit for signed-receipts/v1, with test vectors for implementers: https://councilof.ai/spec/signed-receipts/v1/conformance/
- Downloads last month
- 145