Conda update breaks cuda?

After doing a conda update, I am now getting:

/home/bharland/anaconda3/envs/openmm-torch/lib/python3.13/site-packages/torch/cuda/__init__.py:174: UserWarning: CUDA initialization: CUDA driver initialization failed, you might not have a CUDA gpu. (Triggered internally at /home/conda/feedstock_root/build_artifacts/libtorch_1752524775208/work/c10/cuda/CUDAFunctions.cpp:109.)

and

torch.cuda.is_available()

gives False. I created this environment today after my previous working environment broke, also on an update. Any advice is much appreciated!

The details of the update are

(torch-gpu) bharland@bigbear:~$ conda update --all
2 channel Terms of Service accepted                                             
Channels:                                                                       
 - conda-forge                                                                  
 - defaults                                                                     
Platform: linux-64                                                              
Collecting package metadata (repodata.json): done                               
Solving environment: done                                                       

## Package Plan ##

  environment location: /home/bharland/anaconda3/envs/torch-gpu


The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    numexpr-2.10.2             |py310hdb6e06b_100         169 KB  conda-forge
    pandas-2.3.1               |  py310h0158d43_0        12.0 MB  conda-forge
    ------------------------------------------------------------
                                           Total:        12.2 MB

The following NEW packages will be INSTALLED:

  dav1d              conda-forge/linux-64::dav1d-1.2.1-hd590300_0 
  fsspec             conda-forge/noarch::fsspec-2025.7.0-pyhd8ed1ab_0 
  kaldi              conda-forge/linux-64::kaldi-5.5.1112-cpu_h7360626_9 
  libabseil          conda-forge/linux-64::libabseil-20240722.0-cxx17_hbbce691_4 
  libavif16          conda-forge/linux-64::libavif16-1.3.0-h766b0b6_0 
  libde265           conda-forge/linux-64::libde265-1.0.15-h00ab1b0_0 
  libheif            conda-forge/linux-64::libheif-1.19.7-gpl_hc18d805_100 
  liblapacke         conda-forge/linux-64::liblapacke-3.9.0-32_he2f377e_openblas 
  libopenblas        conda-forge/linux-64::libopenblas-0.3.30-pthreads_h94d23a6_1 
  libprotobuf        conda-forge/linux-64::libprotobuf-5.28.3-h6128344_1 
  libtorch           conda-forge/linux-64::libtorch-2.6.0-cpu_generic_haed06de_0 
  libuv              conda-forge/linux-64::libuv-1.51.0-hb9d3cd8_0 
  nomkl              conda-forge/noarch::nomkl-1.0-h5ca1d4c_0 
  openfst            conda-forge/linux-64::openfst-1.8.3-h84d6215_3 
  optree             conda-forge/linux-64::optree-0.17.0-py310h03d9f68_0 
  pybind11           conda-forge/noarch::pybind11-3.0.0-pyh9380348_1 
  pybind11-global    conda-forge/noarch::pybind11-global-3.0.0-pyhf748d72_1 
  rav1e              conda-forge/linux-64::rav1e-0.7.1-h8fae777_3 
  sleef              conda-forge/linux-64::sleef-3.8-h1b44611_0 
  torchvision-extra~ conda-forge/linux-64::torchvision-extra-decoders-0.0.2-py310h9a3ef1b_2 

The following packages will be REMOVED:

  blas-1.0-mkl
  ffmpeg-4.4.2-gpl_hdf48244_113
  gnutls-3.7.9-hb077bed_0
  intel-openmp-2022.0.1-h06a4308_3633
  lame-3.100-h166bdaf_1003
  libasprintf-0.25.1-h3f43e3d_1
  libgettextpo-0.25.1-h3f43e3d_1
  libidn2-2.3.8-ha4ef2c3_0
  libtasn1-4.20.0-hb9d3cd8_0
  libunistring-0.9.10-h7f98852_0
  libva-2.22.0-h4f16b4b_2
  libvpx-1.13.1-h59595ed_0
  libwebp-1.6.0-h9635ea4_0
  llvm-openmp-15.0.7-h0cdce71_0
  mkl-2022.1.0-hc2b9512_224
  nettle-3.9.1-h7ab15ed_0
  openh264-2.3.1-hcb278e6_2
  p11-kit-0.24.1-hc5aa10d_0
  pytorch-mutex-1.0-cuda
  torchtriton-3.1.0-py310
  wayland-protocols-1.45-hd8ed1ab_0
  x264-1!164.3095-h166bdaf_2

The following packages will be UPDATED:

  aom                                      3.6.1-h59595ed_0 --> 3.9.1-hac33072_0 
  libblas                              3.9.0-16_linux64_mkl --> 3.9.0-32_h59b9bed_openblas 
  libcblas                             3.9.0-16_linux64_mkl --> 3.9.0-32_he106b2a_openblas 
  liblapack                            3.9.0-16_linux64_mkl --> 3.9.0-32_h7ac8fdf_openblas 
  numexpr                             2.7.3-py310hb5077e9_1 --> 2.10.2-py310hdb6e06b_100 
  pandas                              2.2.3-py310h5eaa309_1 --> 2.3.1-py310h0158d43_0 
  pytorch            pytorch::pytorch-2.5.1-py3.10_cuda12.~ --> conda-forge::pytorch-2.6.0-cpu_generic_py310_hb0ad038_0 
  pytz                                  2024.1-pyhd8ed1ab_0 --> 2025.2-pyhd8ed1ab_0 
  svt-av1                                  1.4.1-hcb278e6_0 --> 3.0.2-h5888daf_0 
  torchaudio          pytorch::torchaudio-2.5.1-py310_cu121 --> conda-forge::torchaudio-2.7.0-cpu_py310h9d13ef8_0 
  torchvision        pytorch::torchvision-0.20.1-py310_cu1~ --> conda-forge::torchvision-0.21.0-cpu_py310_h91c00cc_1 


Proceed ([y]/n)? y


Downloading and Extracting Packages:
                                                                                
Preparing transaction: done                                                     
Verifying transaction: done
Executing transaction: done